Multi-equipment cooperative optimization method and system of ground source heat pump air conditioning system

Mode of execution

It is to, be noted that both the following detailed, description and the following detailed description are. exemplary and are intended, to provide further explanation of the disclosure that all technical and scientific terms used herein have the same meaning as commonly understood by one of. ordinary skill in the art to which this disclosure belongs.

It is to, be noted that the terminology used herein is for the purpose of describing particular, embodiments only and is not intended to be limiting to the. exemplary embodiments of the " present, disclosure as used herein unless, the context clearly dictates otherwise, "/ " ".

An embodiment 1: the present invention.

The multi-1 device, cooperative optimization method for 1 the ground source heat pump air-conditioning system according to the embodiment, of: the present disclosure as shown in FIGS.

The system total heat exchange amount and the total energy consumption of, the system are optimized by the system total heat exchange amount and the total, system energy consumption of the system based on the system total heat exchange amount of the system total heat exchange amount and the; total system energy consumption.

The controller of the air conditioning system key equipment system 2 of the, ground source heat pump comprises a controller of the heat source machine set, consisting of a buried heat, N exchanger and a heat pump unit, and a controller of the end, unit N.

The optimal energy-saving control strategy of the energy-saving control model is obtained by solving, the constructed energy-saving control. model optimal solution by adopting a distributed optimization method based on group intelligence.

A concrete construction method of the key device II-type fuzzy model of the ground source heat pump air-conditioning system according, to the embodiment is: as follows.

A multi,input and multi-output heat source unit two- mode fuzzy model is established to a model structure and a parameter,constrained transformation unit, and a boundary constraint condition is taken into consideration for the, input variable set of the two- type fuzzy model of the heat source unit to output variable sets and parameters thereof.

The data- driven two-mode fuzzy rule self-organizing method: based on the two-mode fuzzy model of the heat source unit considering boundary, constraint conditions is used for establishing the constraint-constrained data-driven; self-organizing two-form fuzzy rule base.

In the, parameter self-learning optimization training process of: the heat source unit two-type fuzzy model, the parameter self-learning optimization problem of the heat source unit two-mode fuzzy; rule is obtained by the parameter self-learning optimization training process of the, heat source unit two-type fuzzy model.

In a step-a- step

1)The two-type fuzzy system t, input f, parameter p, set, which v constitutes the two,type fuzzy system input parameter set, forms a multi HE-input EC and multi- output heat source unit two.mode fuzzy model based on the input variables of the. water inlet and the return water temperature difference of the input variables.

2)Both the model input parameter and the output parameter adopt, a Gaussian model, and the water inlet and the return. water temperature difference of one, of the input variables t are in the fuzzy set, and the membership degree when the water inlet and the return water temperature difference are in the fuzzy set is an interval value. The field membership degree can better deal with the strong uncertainty caused by the non-continuous consistency of the peripheral geological environment of, the buried heat exchanger, and facilitates the. modeling of data with various uncertainties.

In order to establish the, model, it is necessary to determine the parameter water inflow and return water temperature difference minimum values according to the constraint conditions and the information in t the data._min Water, inlet and return water temperature difference t maximum value device_max The average value of the Gaussian type m membership function and the variance δ may be determined by using the Gaussian type, that: is, (m in the present, embodiment, by m using the Gaussian model as the average value of the membership function and the variance δ.₁ , δ₁ ) and (m₂ , δ₂ When), the 2 water inlet, and the return water temperature difference corresponding to the water, inlet and the return water temperature difference are 1 as, shown in the figure, the two m mean values corresponding to the water inlet and the return water temperature difference are in the range of '', respectively.₁ Sum-up table m₂ The, corresponding two variances are δ (δ), respectively.₁ m₁ And mean delta₂ .

3)Water inlet and return water temperature difference boundary values between the (inlet water) and the return (water temperature) difference limit value between the. inlet water and the return water temperature difference limit value (between the 70 inlet -120 water) and the return water temperature difference (limit/value) between the inlet water and the return water temperature difference. limit value in the 80 heat, source unit in the heat source unit 7 °C 12 °C; 3 °C 6 °C . t._min ≤t_i ≤t_max , (t_min ＝3,t_max ＝12).

4)In the actual system, the - water intake and return t ∈ T, water f ∈ F, temperature difference unit_@ valve open p ∈ P, unit_@ valve v ∈ V- open unitz @ valve open, unit_@ are abstracted as a unitz t ∈ T ∈ θ._T , f ∈ F ∈ θ_F , p ∈ P ∈ θ_P , v ∈ V ∈ θ_V The, input variable boundary constraint set θ & Rx FF1D; {θ, } is constructed and constructed._T , θ_F , θ_P , θ_V Boundary}.condition-to-parameter-pair parameters for data-by m-data₁ , σ₁ , m₂ , σ₂ Structures that constitute the structure are referred to as, structural constraint sets ω & Rx FF1D; and T m_i , σ_i },(i＝1,2).

Step II.

1)Considering various input variables of 4 the heat source (unit, the relation between, the, inlet, water and the) return 2 water temperature difference (of the water return difference of) the water return. difference between the system energy consumption and the model structure and the, parameter constraint on the basis of the model structure and the parameter constraint: is established on the basis of the boundary conditions.

Of: the model structure and the parameter is initialized by the boundary constraint condition, and the correlation between the data and the rule is (set) by the. match degree evaluation index of the rule between the data and the rule, and the threshold value.

The rule-generating set- up method uses a rough set method to, preliminarily build a rule base: for example, a rule from data.

Where. A two-type fuzzy set of fuzzy components T, under F, boundary P constraint conditions are respectively V input variable water inflow and return water temperature difference of the; water return difference of the water temperature of the water return difference between the valve opening degree and the valve opening degree. sum Two-mode fuzzy suffix set for HE outputting variable EC heat exchange amount and energy. consumption

Regular AD-hoc network:

If the degree of match: satisfies the threshold, requirement, the rule parameters of the rule are adjusted. to satisfy the requirement of the, threshold requirement and the parameters of the rule, are adjusted, to satisfy, the requirement of the threshold.

The rule- consolidating the correlation evaluation and merging threshold values between, the rules, and when the correlation between the, rules satisfies the merge, threshold, the rule undertakes the rule and the. rule parameters are adjusted within the parameter constraint index at the same time.

If the rule: meets the splitting condition, the rule, splitting condition and the splitting rule are that the rule needs, to be, split according to the rules under the condition. that the splitting condition is satisfied, and the rule merging is not formed by the rule after the disassembly.

The rule deletion: unit performs the trimming for the initialization phase according to the set index, to. remove the redundancy rule.

In step three.

The constraint set θ & Rx FF1D; {θ, which is constructed based on the step one._T , θ_F , θ_P , θ_V Y} & omega &E * FF1D; T m_i , σ_i The}, optimization of the parameters of the built-up two-model fuzzy model by using the data is assumed. to have a set of (x N data of one's group.¹ , y¹ ),(x^N , y^N Wherein A), is as follows. The input data of the two-mode, fuzzy model, namely, T, the F, inflow water and the return water temperature difference, the flow rate of the flow rate of the P flow rate of the valve and V; the opening degree of the valve. The model output data includes the, heat exchange and. the energy consumption.

The objective function for constructing the two-model fuzzy model parameter training is as follows. sum Each model relates to a model x.^k Output of an output of a vehicle .J_HE A (θ, ω)/A J converter_EC The (θ, ω) functions of the heat exchange objective function and the energy consumption target are respectively taken. as the heat exchange objective function.

In-flight vehicle, J_HE A (t,f,p,v, θ)/A J converter_EC Each (t,f,p,v, θ) of the input variables and the return water temperature difference unit_@ valve open unit_@ valve open unit_@ and θ &E p ∈ P, FF1D; {θ, respectively, are the input variables._T , θ_F , θ_P , θ_V Total} heat exchange amount of system and total energy consumption; of system sum The heat exchange amount and the energy consumption of the second type i(i＝1,2,m) of fuzzy model of the second key device, namely i the first device, of the second, λ type of fuzzy model of the second key device under the input variable and the constraint condition set respectively._i The heat exchange i efficiency of the first key device is equal, m to the number of key devices. in the system.

The system total heat exchange amount and the total system energy consumption of the system are optimized by the system total heat exchange amount, and the total energy consumption of the system, and the: multi-objective optimization problem of the following band constraint is solved.

s.t.t ∈ θ_T , f ∈ θ_F , p ∈ θ_P , v ∈ θ_V

Θ is that the input variable boundary constraint pre; Ω-set range is in the range. of a structure constraint.

4)Since the output of the built-up two-mode fuzzy model and the parameters of membership functions in the rules, are non-linear, the optimization problem is a band-constrained non- linear multi-objective (optimization problem,).

This embodiment transforms the built-up two-mode fuzzy parameters to the, constrained multi-objective optimization problem based on the parametric optimization problem of the constructed two- mode fuzzy model as non-linear between the output of the built-in blur model and the parameters of the, membership function in the rule, which is a band-constrained nonlinear multi- objective optimization problem.

, Two, methods can be employed for obtaining the optimal: parameter (s).

If A) the objective function is present with respect to the partial derivative, of the parameter and is easy to determine, the constrained multi-objective optimization problem is converted into unconstrained, multi-objective optimization problem by a; penalty function, and then a gradient descent method can be adopted to solve the problem.

If B) the deviation derivative of the objective function with respect to the parameter, does not exist or is extremely difficult to be, found, it is optimized by using a nonlinear optimization algorithm (such as a particle swarm algorithm, a particle swarm algorithm, a neural network, a) neural network . a deep network, and the like.

The group intelligence energy-N saving control model based on, the thermal load demand response of the heat source unit at the moment of time is established by the: heat source unit, and the group intelligent energy-saving control model based on the thermal load demand response of the heat source unit is set.

In-flight vehicle, ω_i The energy consumption and the i(i＝1,2,N) thermal load weight factor of the first heat exchange unit and the thermal load weight factor are in a range [0,1],p, of from one._i The (τ) electric energy i, P conversion efficiency of the first heat exchange unit at the time of the time (τ) of the first heat exchange unit is higher than that of the second heat exchange unit._i^HSU The (τ) electric energy, i consumption of the first heat exchange unit is increased by a factor of from zero. Energy consumption reference mean, HE value of heat source unit_i The (τ) heat exchange i amount at the time of the first heat, HE source unit at the time (τ) of the first heat source unit is in the range of the time interval (τ)._B The heat exchange benchmark value of the heat .P{ ·} source unit indicates that the constraint condition is, β met._HE Sum-up table β_P Threshold, HE value set for setting_max sum Maximum heat exchange amount and maximum electric energy consumption. amount

Method for solving optimal solution of energy-saving control model constructed by adopting distributed optimization method f based on group intelligence^* The specific aspect: ratio thereof is in the range of from about sup .

In this engineering, as, an example of a, heat source unit, the individual devices, of the unit are optimally interpreted as the total energy consumption and the total heat exchange amount is. the maximum as the total energy consumption.

In-flight vehicle, f_i (x_i . Sup .7). g_i (x_i A/A ratio thereof) is in a range of from HE about_i (τ)≤HE_max .

The distributed optimization method of the group intelligence is specifically characterized, in that: the group intelligent distributed optimization method comprises the: following steps.

The (1) particle swarm optimization of, the particle swarm includes the position and; the velocity of the particle.

The (2) degree of adaptation of each particle is calculated by a factor; of from.

Updating the (3) position and velocity of the particles according to the degree; of adaptation

The update formula of (4) the position and the speed thereof is v as follows._ij (t+1)＝v_ij (t)+c₁ r₁ (t)[p_ij (t)-x_ij (t)]+c₂ r₂ [p_gi (t)-x_ij (t)]x_ij (t+1)＝x_ij (t)+v_ij Multi (t+1)-layer chain c saw blade₁ , c₂ It is also an acceleration constant of, the learning factor of at least one ;r of the learning factors.₁ , r₂ Homogeneous random numbers that are in the 0 range 1 of between-and-to, v-zero_ij The velocity of the particles ranges from the speed range; of the user setting the particles.

Whether (5) or not the iteration number reaches the iteration number, is reached or the, most solution condition is reached is 2 ended, and if not, the return is continued to proceed to the search by the search process.

The end of (6) the. procedure.

An embodiment 2: the present invention.

The embodiment of the invention 2 provides a multi-device cooperative optimization system for a ground source heat pump air conditioning system and a system comprising: the same.

The system total heat exchange amount and, the total energy: consumption of the system are optimized by the system total heat, exchange amount and the total system energy consumption of the system based on the, system total heat exchange amount of the system total heat exchange amount and the total system energy consumption .

The model building module, is configured to: establish a group intelligent energy-saving control model based on a thermal, load demand response based on a controller corresponding to a plurality of; heat exchange units of the heat source unit.

The optimal energy- saving control: strategy of the energy-saving control model is obtained by solving the constructed energy-saving, control model optimal solution by a. distributed optimization method based on group intelligence.

The total heat exchange capacity of the system established by the multi-objective optimization problem establishment module and the total energy consumption model of the system are as: follows.

System total heat exchange amount and total energy consumption of system can be abstracted into: device

, The problem of multi-objective optimization with constraint is: solved.

s.t.t ∈ θ_T , f ∈ θ_F , p ∈ θ_P , v ∈ θ_V

In-flight vehicle, J_HE A (t,f,p,v, θ)/A J converter_EC This unit_@ and valve open unit_@ and valve open unit_@ and θ t ∈ T, and θ f ∈ F,E p ∈ P * FF1D;{ θ, v ∈ V- respectively, in the input variables._T , θ_F , θ_P , θ_V Total}- heat exchange amount of system and total energy consumption. of system sum The heat exchange amount and the energy consumption of the second type i(i＝1,2,m) of fuzzy model of the second key device (, namely i the first device, of the second), λ type of fuzzy model of the second key device under the input variable and the constraint condition set respectively._i The heat exchange i efficiency of the first key device is equal, m to the number of key devices. in the system.

As possible some implementations, the heat, source unit is composed N of three heat exchange units, and the group intelligence energy-saving control model based on the thermal load demand response of the: heat source unit at the time of the τ can be set.

Method for solving optimal solution of energy-saving control model constructed by adopting distributed optimization method f based on group intelligence^* The specific aspect: ratio thereof is in the range of from about sup .

In-flight vehicle, ω_i The energy consumption and the i(i＝1,2,N) thermal load weight factor of the first heat exchange unit and the thermal load weight factor are in a range [0,1],p, of from one._i The (τ) electric energy i, P conversion efficiency of the first heat exchange unit at the time of the time (τ) of the first heat exchange unit is higher than that of the second heat exchange unit._i^HSU The (τ) electric energy, i consumption of the first heat exchange unit is increased by a factor of from zero. Energy consumption reference mean, HE value of heat source unit_i The (τ) heat exchange i amount at the time of the first heat, HE source unit at the time (τ) of the first heat source unit is in the range of the time interval (τ)._B The heat exchange benchmark value of the heat .P{ ·} source unit indicates that the constraint condition is, β met._HE Sum-up table β_P Threshold, HE value set for setting_max sum Maximum heat exchange amount and maximum electric energy consumption. amount

An embodiment 3: the present invention.

An embodiment of the present 3 disclosure provides a method for implementing the, multi-device cooperative, optimization method of the ground source heat pump air conditioning system 1 according to the first embodiment of the present disclosure when the program is executed. by the processor when the program is executed by the processor.

An embodiment 4: the present invention.

In one embodiment of the 4 present disclosure, the method includes, the steps, of: performing the program when the processor executes the program, the processor, executing the program, and executing the program on the processor 1, and the processor executes the program in the ground source heat pump air. conditioning system multi-device cooperative optimization method.

The above description is merely a preferred embodiment of the present disclosure, and is not intended to, limit the present disclosure to those skilled, in the art that various modifications and variations of. the present disclosure may be included within the, scope of protection of the, present disclosure, without departing, from the spirit and principle of the disclosure.