Electric power trading information publishing risk identification method

1. A power trading information issuing risk identification method, characterized in that the method steps are as follows:

Step 1: points defining risk, risk-point is stored in the database in the form of a, so in the implementation of the risk when the detection algorithm can be directly taken out from the database risk point, according to the attribute of the risk point carries out recognition algorithm;

Step (11) to determine the risk to be detected, the risk of the risk definition-point, point-to-risk adding risk in storehouse , the system firstly it is judged whether there is the risk of the same point, if there is the risk of the same point, to re-input to the risk of each other;

Step (12): if there are no duplicate name, the system will risk point in adding risk storehouse , and requires an input of the description information of the point of this risk, risk type described including the main information, risk levels, risk recognition priority, risk recognition start time, risk recognition activation interval time, relates to database table, fields of the table, the expected value of the word, to risk index, the risk of the field of influence factor, alarm level risk, risk alarm notifiers;

Step (13): information input after described, the system will describe information quantization, that is, into a suitable mode of storing the database, store in a database;

Step (14): then the system requires an input of risk recognition algorithm, algorithm specific preparation process in the step 2 on in, the risk is received by the system after the recognition algorithm will carry out an analog operation, this operation is mainly whether to correctly execute the test algorithm, whether right-end, the analog algorithm is not changed in the course of running any information in the database, only the algorithm is risk recognition module to call its operation will only affect the information of the database table;

Steps (15): after the verify correct, the algorithm to store in a database, and the risk associated with the corresponding point;

Steps 2: risk definition of the recognition algorithm, according to the definition of the characteristics of the risk point, defining the risk point recognition algorithm;

Steps (21): define storing risk point quantization information variable, is taken out from a risk storehouse point of this risk in the quantization information and to the corresponding variable assignment;

Steps (22): from these quantization information parsing point of this risk and related fields associated with the database table;

Step (23): entering a circulation, taken out from the database table the value of the relevant field, and the field stored in the risk storehouse or comparing the expected value of the calculation, the field obtained by calculation according to the results of the risk index, and then multiplied by the fields of the risk factor, the risk values, the field, the value accumulated to risk early warning-level, processing after the end of the same a the operation of the field, until the processing end all of the field;

Steps (24): the warning-level and the risk of the risk of each field is stored to the result base table value, subsequent to the risk analysis;

Steps 3: risk recognition, the risk of the point when the identification time, calling and executing risk point identification algorithm, risks after the end point identification algorithm implementation, call release the module execution result of the risk, risk early warning issued to the relevant personnel;

Steps (31): system for each added to the risk of risk library point set up a timer, each time the risk of the point of time when the detection of the, system will be the risk point by adding risk identification queue, this is a risk point identification of prioritizing the risk of a priority queue;

Step (32): risk point into the queue will measure the risk identification module is dormant, if the identification module is dormant, activate the risk identification module, if the risk identification module is not dormant, any operation is not performed;

Step (33): risk module when in the dormant state, continuously from a risk identification queue implementation risk-point and then taken out, the implementation process, based on risk point out risk recognition algorithm ID, call algorithm running engine, algorithm engine loading risk recognition algorithm and operation, automatic algorithm will run to the end the operation result information in the recognition result by adding risk, and withdraw from the algorithm engine operation, the control power to the risk identification module;

Step (34): risk identification module in the recognition result taking risk risk early-warning-level, and with the risk-point is compared to the alarm level, if the risk of early warning-level edge, the risk of early warning information to the risk alarm notifiers, if the risk of early warning level is low, the risk should be on the module automatically processing risk, risk identification result according to the processing procedure of the data in the library of the fine adjustment of the information to be published;

Step (35): after the end of treatment of the risk identification module from a risk identification queue a to be detected of the risk point, if we can get the risk point, implementation of the above-mentioned risk identification process, if no risk point to get the, risk recognition module and enters the dormant state.