Method for identifying and analyzing abnormal bidding behavior of electric power bidding

Abstract

The invention discloses a method for identifying and analyzing an abnormal bidding behavior of power bidding, which automatically identifies and outputs a comprehensive risk index by respectively performing parameter deviation analysis, working condition mismatch analysis and quotation abnormality analysis, realizes accurate matching of parameters and an actual engineering environment, avoids subjective deviation caused by manual auditing, and improves the efficiency of bidding. And meanwhile, an automatic analysis process greatly improves the anomaly recognition efficiency, and a multi-dimensional scoring fusion mechanism enhances the comprehensiveness and accuracy of risk assessment. According to the invention, through the pre-constructed working condition-equipment parameter correction mapping rule base, the corresponding to-be-adjusted type equipment is identified based on the actual working condition of the project location corresponding to the current bid invitation project, the to-be-adjusted equipment can be accurately identified based on the actual working condition of the project location, and through the working condition adaptation rule extracted through historical maintenance data, the equipment adjustment efficiency is improved. Matching equipment parameters with actual environment requirements; working condition mismatching risks are found in advance, equipment faults caused by environment discomfort are avoided, and the operation reliability of engineering equipment is improved.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of abnormal bidding behavior identification, and relates to an electric power bidding abnormal bidding behavior identification analysis method. BACKGROUND

[0002] In the electric power engineering bidding process, abnormal bidding behaviors of bidding units, such as non-compliance of equipment parameters, abnormal bidding, and insufficient adaptability of working conditions, may cause engineering quality problems, progress delays, and cost overruns. Therefore, the accurate identification of abnormal bidding behaviors in electric power bidding is of great significance.

[0003] The traditional technical solution usually uses manual auditing to identify abnormal bidding behaviors. Manual checking is slow and difficult to handle large quantities of bids. It is also affected by experience differences, leading to inconsistent standards. It is easy to overlook complex and hidden abnormal patterns, and it requires continuous investment in professional manpower, which is costly. Overall, it is difficult to meet the needs of accurate and efficient risk identification.

[0004] The traditional technical solution lacks adaptability in identifying abnormal equipment parameters based on actual working conditions, which may lead to mismatch between equipment parameters and working conditions in the project location, causing equipment failures in actual operation, affecting engineering quality and progress, and failing to identify potential risks caused by working condition mismatch in advance, increasing the cost of post-project maintenance and safety hazards. SUMMARY

[0005] In view of the above problems in the background art, an electric power bidding abnormal bidding behavior identification analysis method is proposed.

[0006] The purpose of the application can be achieved by the following technical solution: an electric power bidding abnormal bidding behavior identification analysis method, comprising: device list construction: obtaining the required number of each type of device, parameters and prices from each bidding document, and organizing the device list.

[0007] Parameter deviation analysis: match each parameter of each type of device with the pre-provided standard value and analyze the correlation between the internal parameters of the device to identify each abnormal type of device and output the parameter deviation score.

[0008] Working condition mismatch analysis: obtain the actual working condition of the project location corresponding to the current bidding project, identify the type of device to be adjusted corresponding to the current bidding project based on the pre-constructed working condition-device parameter correction mapping rule library, and further output the working condition mismatch score.

[0009] Bidding anomaly analysis: compare the prices of each type of device in the device list with the reference market prices of the components corresponding to each device at the current bidding time, and identify abnormal bidding devices based on the horizontal comparison results of the prices, and output the bidding anomaly score.

[0010] The comprehensive risk output: based on the parameter deviation score, the working condition mismatch score and the bid abnormality score, output the comprehensive risk index corresponding to each bid file, and determine whether each bid file has abnormal bidding behavior.

[0011] Compared with the prior art, the beneficial effects of the present application are as follows: (1) The present application automatically identifies and outputs the comprehensive risk index by performing parameter deviation analysis, working condition mismatch analysis and bid abnormality analysis, respectively, realizes accurate matching of parameters and actual engineering environment, avoids subjective deviation caused by manual review, and at the same time, the automatic analysis process greatly improves the abnormal identification efficiency, the multi-dimensional score fusion mechanism enhances the comprehensiveness and accuracy of risk assessment, and provides a scientific and objective decision basis for power bidding.

[0012] (2) The present application can accurately identify the equipment to be adjusted based on the actual working condition of the bidding engineering site by using the pre-constructed working condition-equipment parameter correction mapping rule library, and based on the actual working condition of the bidding engineering site, the corresponding equipment to be adjusted is identified, the working condition adaptation rule extracted from the historical maintenance data is used to match the equipment parameters with the actual environmental requirements, the working condition mismatch risk is found in advance, the equipment failure caused by environmental inadaptability is avoided, the operation reliability of engineering equipment is improved, and the maintenance cost and safety hazards are reduced. BRIEF DESCRIPTION OF DRAWINGS

[0013] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor.

[0014] Figure 1 The method steps of the present application are illustrated in the schematic diagram.

[0015] Figure 2 The judgment flowchart of whether the bid file corresponding to an embodiment provided by the present application has abnormal bidding behavior. DETAILED DESCRIPTION

[0016] The technical solutions in the embodiments of the present application will be described in detail below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0017] Please refer to Figure 1As shown, the present application provides an electric power bidding abnormal bidding behavior identification analysis method, comprising: device list construction: obtaining the required number of each type of device, parameters and prices from each bidding file, and arranging to obtain the device list.

[0018] For example, the device includes but is not limited to transformer, circuit breaker, lightning arrester, etc., and different models of the same device are also regarded as different device types.

[0019] It should be explained that the device parameters refer to the specific technical performance indicators possessed by each type of device, for example, in the electric power bidding scene, the rated voltage of the transformer, the short-circuit current resistance value, the short-circuit current resistance value of the circuit breaker, etc. These parameters are used to measure whether the device meets the bidding requirements and the actual working condition requirements, and are important basis for device list construction and subsequent parameter deviation analysis, working condition mismatch analysis, etc.

[0020] Parameter deviation analysis: match each parameter of each type of device with the pre-provided standard value and analyze the device internal parameter correlation to identify each abnormal type of device and output the parameter deviation score.

[0021] In a preferred embodiment of the present application, the specific analysis method for identifying each abnormal type of device is as follows: the absolute difference value of each parameter of each type of device is calculated respectively with the pre-provided standard value to obtain the relative deviation of each parameter of each type of device, and then the relative deviation degree of each parameter of each type of device is calculated by ratio to obtain the relative deviation degree of each parameter of each type of device.

[0022] It should be noted that the above standard value refers to the reference value for measuring the compliance of the device parameters, and its source is the technical requirements of the bidding party.

[0023] The relative deviation degrees of each parameter of each type of device are compared, and the maximum relative deviation degree is selected as the relative deviation degree of the corresponding parameter of each type of device, and then compared with the preset relative deviation degree threshold.

[0024] It should be noted that the reason for selecting the maximum relative deviation degree as the relative deviation degree of the corresponding parameter of each type of device: there may be multiple parameter deviations for the device, but the maximum deviation degree usually reflects the parameter that is most inconsistent with the standard, which is a key factor affecting the performance of the device. By focusing on the maximum deviation, the most dangerous parameter abnormality can be quickly located. In the electric power device bidding, a key parameter such as voltage level and insulation performance may directly lead to the device failing to meet the engineering requirements, and even cause safety hazards, so using the maximum deviation as the basis for judgment is more in line with the actual risk control requirements.

[0025] Based on the correlation logic in the pre-set parameter correlation list, the related parameters of each type of device are extracted.

[0026] It should be noted that in the parameter deviation analysis, in addition to the deviation of a single parameter from the standard value, the correlation between the internal parameters of the equipment also needs to be considered. For example, the rated voltage and rated current of the power equipment may have a fixed logical relationship, such as a power formula correlation. If a parameter deviates from the standard value and breaks this correlation, it may indicate that the equipment has an anomaly.

[0027] It should be explained that the preset parameter correlation list is a pre-established rule library that records the logical relationships between different parameters in various types of equipment, such as mathematical formulas, engineering specification requirements, etc. For example, the correlation logic in the parameter correlation list can be: 1. In a direct current equipment or a purely resistive load: under rated operating conditions, the rated power is equal to the product of the rated voltage and the rated current. 2. In an alternating current equipment: under rated operating conditions, when the equipment's nominal rated power is active power, the rated power is less than or equal to the product of the rated voltage and the rated current; when the equipment's nominal rated power is apparent power, the rated power is equal to the product of the rated voltage and the rated current.

[0028] The relevant parameters of the various types of equipment are matched and analyzed according to the correlation logic to determine whether there is any correlation logic abnormality in the relevant parameters of the various types of equipment.

[0029] It should be noted that the identification of correlation logic abnormality has the following significance: 1. Revealing technical parameter fraud: bidders may intentionally adjust other parameters to meet a single parameter standard value, resulting in a logical contradiction between parameters, such as intentionally reducing the power parameter to meet the price requirement, but ignoring the correlation between voltage and current. This analysis can identify such implicit abnormalities. 2. Supplementing the shortcomings of single parameter analysis: even if the relative deviation degree of a single parameter does not exceed the threshold, if the logical relationship between the parameters is broken, the equipment can still be determined to be abnormal, avoiding missed judgments.

[0030] When the parameter relative deviation degree is greater than the preset relative deviation degree threshold or there is any correlation logic abnormality, the corresponding type of equipment is determined to be an abnormal type of equipment.

[0031] In a preferred embodiment of the present application, the specific analysis method of the parameter deviation score is as follows: based on the equipment list, the number of devices corresponding to each abnormal type of equipment and the price are obtained.

[0032] The number of devices corresponding to each abnormal type of equipment and the price are multiplied to obtain the total price of each abnormal type of equipment.

[0033] The number of devices corresponding to each type of equipment and the price are multiplied to obtain the total price of each type of equipment.

[0034] The total price of each abnormal type of equipment and the total price of each type of equipment are calculated by ratio to obtain the parameter deviation score.

[0035] It should be noted that the parameter deviation score realizes the leap from technical anomaly identification to economic risk assessment by the economic impact of the price proportion parameter deviation, and is the key link connecting the equipment parameter compliance and the bid cost, providing an evaluation dimension with technicality and economy for the tenderer to screen compliant bid documents.

[0036] Working condition mismatch analysis: Obtain the actual working condition of the project site corresponding to the current bidding project, identify the type of equipment to be adjusted corresponding to the current bidding project based on the pre-constructed working condition-equipment parameter correction mapping rule library, and further output the working condition mismatch score.

[0037] It should be noted that the working condition of the power engineering site, such as climate, geological conditions, will directly affect the equipment parameter demand, such as corrosion prevention in coastal areas and low temperature resistance in high-cold areas. The rule library dynamically binds the working condition and parameter demand through historical data statistics to avoid one-size-fits-all parameter standards.

[0038] In a preferred embodiment of the present application, the specific construction process of the working condition-equipment parameter correction mapping rule library is as follows: based on historical equipment maintenance records, obtain the equipment type and parameters to be optimized corresponding to each equipment maintenance operation, count each parameter to be optimized corresponding to each equipment type, sort each parameter to be optimized according to the frequency of occurrence, identify core parameters according to the preset number ratio based on the sorting result, and record other parameters to be optimized as secondary parameters.

[0039] It should be noted that the significance of distinguishing core parameters and secondary parameters is as follows: 1. Focus on key influencing factors: High-frequency parameters are usually the indicators that are most likely to cause performance deviation of equipment under different working conditions, such as the corrosion prevention coating parameters of equipment in coastal areas that are frequently maintained due to environmental impact. Identifying core parameters can improve the efficiency and accuracy of working condition matching analysis.

[0040] 2. Optimize rule library construction cost: The number of core parameters is small, which can concentrate resources for in-depth analysis of the working condition-parameter mapping relationship, avoiding redundancy of the rule library due to too many parameters.

[0041] Based on historical equipment maintenance records, obtain the actual demand values of each type of equipment corresponding to each core parameter in a large number of projects, and obtain the actual working conditions of each project.

[0042] It should be noted that the actual demand value statistics based on historical maintenance records, rather than subjective parameter setting, make the rule library more consistent with engineering practice and reduce the risk of disconnection between theoretical parameters and actual demand. By analyzing the actual demand values of core parameters under different working conditions, the parameter optimization direction of equipment in a specific environment is found. For example: the actual demand value of the corrosion prevention coating thickness of equipment in coastal areas is generally higher than that in inland areas; the oil liquid freezing point parameter of the transformer in high-cold areas needs to be adjusted to a lower value.

[0043] The actual demand values of each type of device corresponding to each core parameter are classified according to the working condition types, and the actual demand values of each core parameter of each type of device corresponding to each working condition are obtained by mean calculation of different actual demand values of the same type of device and the same core parameter.

[0044] The working conditions and the actual demand values of each core parameter of each type of device are one-to-one corresponding, and a working condition-device parameter correction mapping rule library is constructed.

[0045] The working conditions in the above working condition-device parameter correction mapping rule library can be, but are not limited to, one or more of coastal, high-cold and seismic belt.

[0046] It should be noted that the rule library at least includes typical working conditions such as coastal, high-cold, seismic belt, and can also be extended to other working condition types according to actual needs, such as high-temperature and high-humidity areas.

[0047] In a preferred embodiment of the present application, the specific analysis method for identifying the type of device to be adjusted corresponding to the current bidding project is as follows: the actual working condition of the area where the current bidding project is located is obtained, and then matched with the working condition-device parameter correction mapping rule library to obtain the actual demand values of each parameter of each type of device.

[0048] The standard values of each parameter of each type of device corresponding to the current bidding project are compared with the actual demand values to obtain the relative deviation degree of each parameter of each type of device.

[0049] The relative deviation degree is compared with the pre-set relative deviation degree threshold value, and the parameters with a relative deviation degree greater than the relative deviation degree threshold value are recorded as risk parameters.

[0050] It should be noted that the setting of the relative deviation degree threshold value is based on: by analyzing the historical equipment maintenance records, the correlation between the deviation degree of different parameters and the equipment failure and maintenance cost is counted. For example, if the deviation degree of a certain parameter exceeds 15%, the failure rate of the equipment in the high-cold area significantly increases, and then 15% is set as the threshold value under this working condition. The threshold value can be determined based on the statistical distribution of the actual demand values of the parameters corresponding to each working condition in the rule library, such as a combination of mean value and standard deviation, reflecting the parameter fluctuation range of the normal operation of the equipment in most historical projects.

[0051] If any parameter of a certain type of device is a risk parameter, the type of device is marked as a type of device to be adjusted, and each type of device to be adjusted is compared and counted.

[0052] It should be noted that even if other parameters of the device meet the requirements, as long as there is a risk parameter, the device may not be able to operate normally under actual working conditions, such as cable temperature resistance parameters not meeting the requirements in high-cold regions, which may freeze and crack. Therefore, the principle of determining abnormality by any parameter anomaly directly marks the device type as a device to be adjusted to avoid the risk of parameter mismatch comprehensively.

[0053] In a preferred embodiment of the present application, the specific analysis method of the working condition mismatch score is as follows: the actual relative deviation degree of each parameter of each type of device to be adjusted is calculated by comparing each parameter of each type of device to be adjusted with the corresponding actual demand value.

[0054] The average actual relative deviation degree of each type of device to be adjusted is calculated by averaging the actual relative deviation degree of each parameter.

[0055] The type of device to be adjusted with the average actual relative deviation degree greater than the preset relative deviation threshold is recorded as a working condition abnormal device, and the number of working condition abnormal devices and the number of devices to be adjusted are counted to obtain the working condition mismatch score.

[0056] It should be noted that the ratio of the number of working condition abnormal devices to the number of devices to be adjusted is used as the working condition mismatch score, because the average deviation degree is used to filter abnormal devices whose parameters deviate from the working condition requirements as a whole, and then the risk concentration is quantified by the proportion, which can objectively reflect the proportion of devices that are severely incompatible with the working conditions in the bidding devices, avoid misjudgment of single parameter risk, and the numerical index is convenient for the bidding party to make decisions, and can also connect the rule library to prevent batch device failure.

[0057] It should be noted that the present application can accurately identify the type of device to be adjusted based on the actual working conditions of the bidding project site through the pre-constructed working condition-device parameter correction mapping rule library, and based on the actual working conditions of the bidding project site. The working condition adaptation rule extracted from the historical maintenance data can match the device parameters with the actual environmental requirements, discover the working condition mismatch risk in advance, avoid device failure caused by environmental inadaptability, improve the reliability of the engineering equipment, and reduce the maintenance cost and safety hazards.

[0058] Abnormal offer analysis: compare the price of each type of device in the device list with the reference market price of each component corresponding to the device at the current bid evaluation time, and identify abnormal offer devices based on the horizontal comparison result of the price, and output the abnormal offer score.

[0059] In a preferred embodiment of the present application, the specific analysis method for identifying abnormal bidding equipment is as follows: extracting each type of equipment from the equipment list, obtaining the constituent parts and quantity of each type of equipment, obtaining the average market price of the corresponding constituent parts of each type of equipment at the current bidding time, and combining and calculating the price and quantity to obtain the reference market price of each type of equipment at the current bidding time.

[0060] It should be noted that the market price data of each component at the current bidding time can be obtained through industry price database, supplier bidding platform and other channels. The average value of different supplier bids for the same component is taken.

[0061] The reference price of each type of equipment provided in advance is compared with the reference market price of each type of equipment at the current bidding time to calculate the relative deviation value and obtain the dynamic price fluctuation degree of each type of equipment.

[0062] It should be noted that the purpose of dynamic price fluctuation analysis is: 1. Establish a dynamic price benchmark: market prices fluctuate over time, such as the increase in copper prices leading to an increase in winding costs. By calculating the reference value based on the component prices at the current bidding time, the deviation caused by using static standard price is avoided, and the timeliness of the bid analysis is ensured.

[0063] 2. Disassemble the bid rationality verification: verify the rationality of the overall bid price by adding up the component prices to prevent bidders from hiding the overall bid abnormality by artificially lowering the core price and artificially raising the winding price to balance the total price.

[0064] The average bid price of each type of equipment corresponding to each bidding document is calculated, the average bid price is compared with the price marked in the corresponding equipment list, the difference value is calculated, and then the average bid price of each type of equipment is calculated. The ratio of the average bid price of each type of equipment is calculated to obtain the lateral price deviation degree of each type of equipment.

[0065] It should be noted that the purpose of lateral price deviation degree analysis is: 1. Identify group bidding abnormalities: the lateral deviation degree reflects the deviation degree of the bid of a single bidder from the group average, and can find malicious low-price bidding or price string bidding.

[0066] 2. Supplement the market price comparison dimension: unlike the dynamic price fluctuation degree, the lateral deviation degree focuses on the price relationship between bidders, and the combination of the two can comprehensively judge the bid rationality.

[0067] The dynamic price fluctuation degree and the lateral price deviation degree of each type of equipment are compared with the pre-set dynamic price fluctuation degree threshold and the lateral price deviation degree threshold, and the type of equipment with a dynamic price fluctuation degree greater than the dynamic price fluctuation degree threshold or a lateral price deviation degree greater than the lateral price deviation degree threshold is recorded as an abnormal bidding equipment.

[0068] It should be noted that the setting of the dynamic price fluctuation threshold and the transverse price deviation threshold is based on: 1. The dynamic price fluctuation threshold is set according to industry specifications, historical price data and market price trends. Industry specifications clearly define the reasonable range of price fluctuations, such as some engineering construction projects stipulating that material price fluctuations within ±5% are normal. Historical data presents the law of price changes, which is used to determine the long-term fluctuation interval. Market price trends reflect the impact of the current economic environment, such as significant fluctuations in raw material prices and significant changes in supply and demand relationships. According to this, the threshold is adjusted to accurately determine whether the bid price deviates from the market reference price.

[0069] 2. The transverse price deviation threshold is determined based on the bidding price distribution law, which analyzes the dispersion degree of the same type of equipment bid price in a large number of historical bidding data, and calculates statistical quantities such as standard deviation and coefficient of variation. If the standard deviation of the bid price of a certain type of equipment is large, it indicates that the bid price is dispersed, and a higher threshold can be set. If the standard deviation is small, the bid price is concentrated, and the threshold is correspondingly reduced to measure whether the bid price of a single bidder deviates from the average bid price of the group.

[0070] In a preferred embodiment of the present application, the specific analysis method of the bid price abnormality score is as follows: the number of abnormal bid price equipment is counted, and then the ratio calculation is performed with the total number of each type of equipment to obtain the proportion of the number of abnormal risk equipment.

[0071] The average price abnormality degree is calculated by averaging the price abnormality degree of each abnormal bid price equipment.

[0072] The bid price abnormality score is calculated by averaging the proportion of the number of abnormal risk equipment and the average price abnormality degree.

[0073] It should be noted that the proportion of the number of abnormal risk equipment and the average price abnormality degree are averaged to obtain the bid price abnormality score. This method combines the size of abnormal equipment and the degree of price deviation in two dimensions to realize the quantitative evaluation of the rationality of the bid price. Specifically, the proportion of the number of abnormal risk equipment reflects the distribution breadth of the price abnormal equipment in the bid, and the average price abnormality degree reflects the overall amplitude of the bid price deviation from the reference price. The average calculation of the two can take into account the proportion of the number of abnormal equipment and the degree of deviation, avoiding the limitations of a single indicator, and providing comprehensive quantitative basis for judging whether there is malicious bidding, bid rigging and other abnormal behaviors.

[0074] Comprehensive risk output: based on the parameter deviation score, the working condition mismatch score and the bid price abnormality score, the comprehensive risk index corresponding to each bidding document is output to determine whether there is abnormal bidding behavior in each bidding document.

[0075] In a preferred embodiment of the present application, the specific way of outputting the comprehensive risk index corresponding to each bidding document is as follows: the parameter deviation score, the working condition mismatch score and the bid price abnormality score of each bidding document are analyzed.

[0076] The parameter deviation score, the working condition mismatch score and the bid abnormality score are fused and calculated according to weights to obtain a comprehensive risk index corresponding to each bid file.

[0077] It should be noted that the setting of the above weight is based on: collecting historical data under the normal implementation scene of the bidding project, covering the three types of scoring related indicators and the subjective quantitative evaluation index of experts on the comprehensive risk of bidding. The Pearson correlation coefficient of each score and the comprehensive risk evaluation index is calculated, a mapping model of the score and the comprehensive risk is constructed by multiple linear regression, the standardized regression coefficient is extracted as the initial weight, and the weight is normalized to make the sum of the weights equal to 1, so as to determine the weight of the parameter deviation score, the working condition mismatch score and the bid abnormality score in the calculation of the comprehensive risk index.

[0078] In a preferred embodiment of the present application, the specific analysis method for judging whether each bid file has abnormal bidding behavior is as follows: comparing the comprehensive risk index of each bid file with a pre-set comprehensive risk index threshold, and comparing the corresponding parameter deviation score, working condition mismatch score and bid abnormality score with the pre-set parameter deviation score threshold, working condition mismatch score threshold and bid abnormality score threshold respectively.

[0079] It should be noted that the setting of each threshold is based on: referring to industry standards and specifications to clearly define basic requirements such as equipment parameters, working condition adaptation and price; relying on historical project data to statistically analyze the distribution of parameter deviation, working condition mismatch and bid price abnormality of normal bidding, and extracting reasonable ranges; combining expert experience to consider market, technology and abnormal situations; and then accurately anchoring each score threshold according to the characteristics of the project itself, such as importance, size, etc.

[0080] Please refer to Figure 2 As shown in the figure, if the comprehensive risk index is greater than the corresponding threshold or any score is greater than the corresponding threshold, it is determined that the corresponding bid file has abnormal bidding behavior, otherwise it is determined that the corresponding bid file does not have abnormal bidding behavior.

[0081] It should be explained that the logic of the abnormality judgment rule is: when the comprehensive risk index calculated by the weights of the parameter deviation score, the working condition mismatch score and the bid abnormality score exceeds the pre-set threshold, it indicates that the bid file has systematic risk in multiple dimensions such as technical parameters, working condition adaptation and price rationality.

[0082] If any single score exceeds the corresponding threshold, even if the comprehensive index does not exceed the threshold, it is directly determined to be abnormal, so as to avoid that a serious risk in a single dimension is covered up by other indicators.

[0083] Through multi-dimensional comprehensive evaluation and single-dimensional veto, the overall risk of the bid file and the local risk of the key dimension are considered, so as to ensure the comprehensiveness and strictness of the abnormal behavior identification.

[0084] It should be noted that the present application realizes accurate matching of parameters and actual engineering environment by respectively performing parameter deviation analysis, working condition mismatch analysis and quotation abnormality analysis, and then automatically identifying and outputting a comprehensive risk index, avoids subjective deviation caused by manual auditing, and at the same time, the automatic analysis process greatly improves the abnormality identification efficiency, the multi-dimensional score fusion mechanism enhances the comprehensiveness and accuracy of risk assessment, and provides a scientific and objective decision basis for power bidding.

Claims

1. A method for identifying and analyzing abnormal bidding behavior in power tendering, characterized in that, include: Equipment list construction: Obtain the required quantity, parameters and price of each type of equipment from each bidding document, and compile them into an equipment list; Parameter deviation analysis: Match each parameter of each type of equipment with the pre-provided standard values ​​and combine with the internal parameter correlation analysis of the equipment to identify each type of abnormal equipment and output parameter deviation scores; Operating condition mismatch analysis: Obtain the actual operating conditions of the project location corresponding to the current bidding project, identify the type of equipment to be adjusted corresponding to the current bidding project based on the pre-built operating condition-equipment parameter correction mapping rule library, and output the operating condition mismatch score; Price Anomaly Analysis: The prices of each type of equipment in the equipment list are compared with the reference market prices of the corresponding components of each equipment at the current bidding time. Combined with the results of the horizontal price comparison, abnormal price quotations are identified and an abnormal price score is output. Comprehensive risk output: Based on the parameter deviation score, operating condition mismatch score and price anomaly score, output the comprehensive risk index corresponding to each bidding document to determine whether there is any abnormal bidding behavior in each bidding document.

2. The method for identifying and analyzing abnormal bidding behavior in power tendering as described in claim 1, characterized in that: The specific analysis method for identifying each type of anomaly in the device is as follows: The relative deviation of each parameter of each type of equipment is obtained by calculating the absolute difference between each parameter and the pre-provided standard value. Then, the relative deviation degree of each parameter of each type of equipment is obtained by calculating the ratio with the corresponding standard value. The relative deviation of each parameter corresponding to each type of equipment is compared, and the maximum relative deviation is selected as the relative deviation of the parameter corresponding to each type of equipment, and then compared with the preset relative deviation threshold. Based on the pre-defined parameter association list, relevant parameters for each type of device are extracted according to the corresponding association logic. The relevant parameters of each type of equipment are analyzed for correlation logic matching to determine whether there is any abnormal correlation logic in the relevant parameters of each type of equipment. When the relative deviation of the parameters exceeds the preset relative deviation threshold or when any related logical anomaly exists, the corresponding type of device is determined to be an abnormal type of device.

3. The method for identifying and analyzing abnormal bidding behavior in power tendering as described in claim 2, characterized in that: The specific analysis method for the parameter deviation score is as follows: Based on the equipment list, obtain the quantity and price of equipment corresponding to each type of anomaly; The total price of each type of equipment is calculated by multiplying the quantity and price of the equipment corresponding to each type of abnormality. The total price of each type of equipment is calculated by multiplying the quantity of equipment and the price of each type of equipment. The parameter deviation score is obtained by calculating the ratio of the total price of each abnormal type of equipment to the total price of each type of equipment.

4. The method for identifying and analyzing abnormal bidding behavior in power tendering as described in claim 1, characterized in that: The specific construction process of the operating condition-equipment parameter correction mapping rule base is as follows: Based on historical equipment maintenance records, obtain the equipment type and parameters to be optimized for each equipment maintenance operation, count the parameters to be optimized for each equipment type, sort the parameters to be optimized according to their frequency of occurrence, identify core parameters based on the sorting results according to a preset number ratio, and record other parameters to be optimized as secondary parameters. Based on historical equipment maintenance records, the actual required values ​​of each core parameter for each type of equipment in a large number of projects were obtained, and the actual working conditions of each project were also obtained. The actual demand values ​​of each core parameter for each type of equipment are categorized according to the working condition type, and the average value of the actual demand values ​​of the same core parameter for the same type of equipment is calculated to obtain the actual demand values ​​of each core parameter for each type of equipment under each working condition. By mapping the operating conditions one-to-one with the actual required values ​​of each core parameter of each type of equipment, a base of rules for correcting the operating conditions and equipment parameters is constructed. The operating conditions in the above-mentioned operating condition-equipment parameter correction mapping rule base are one or more of the following: coastal, high-altitude cold region, and earthquake zone.

5. The method for identifying and analyzing abnormal bidding behavior in power tendering as described in claim 4, characterized in that: The specific analysis method for identifying the type of equipment to be adjusted corresponding to the current bidding project is as follows: Obtain the actual working conditions in the region where the current bidding project is located, and then match them with the working condition-equipment parameter correction mapping rule library to obtain the actual required values ​​of each parameter for each type of equipment; The relative deviation of each parameter of each type of equipment in the current bidding project is obtained by analyzing the relative deviation between the standard values ​​and the corresponding actual demand values. The relative deviation is compared with a preset relative deviation threshold, and the parameter whose relative deviation is greater than the relative deviation threshold is recorded as a risk parameter. If any parameter of a certain equipment type is a risk parameter, mark the equipment type as an equipment type to be adjusted, and compare and statistically obtain the equipment types to be adjusted.

6. The method for identifying and analyzing abnormal bidding behavior in power tendering as described in claim 5, characterized in that: The specific analysis method for the operating condition mismatch score is as follows: The relative deviation of each parameter of each type of equipment to be adjusted is calculated by comparing it with the corresponding actual demand value to obtain the actual relative deviation of each parameter of each type of equipment to be adjusted. The average actual relative deviation of each parameter is calculated by averaging the actual relative deviations of the equipment to be adjusted. Equipment of the type to be adjusted that has an average actual relative deviation greater than a preset relative deviation threshold is recorded as equipment with abnormal operating conditions. The ratio of the number of equipment with abnormal operating conditions to the number of equipment of the type to be adjusted is calculated to obtain the operating condition mismatch score.

7. The method for identifying and analyzing abnormal bidding behavior in power tendering as described in claim 1, characterized in that: The specific analysis method for identifying abnormal pricing devices is as follows: Extract each type of equipment from the equipment list, obtain the components and quantities of each type of equipment, obtain the average market price of the corresponding components of each type of equipment at the current bidding time, and calculate the reference market price of each type of equipment at the current bidding time by combining the price and quantity. The dynamic price fluctuation of each type of equipment is obtained by calculating the relative deviation between the pre-provided reference prices of each type of equipment and the reference market prices of each type of equipment at the current bid evaluation time. Calculate the average price of each type of equipment in each bidding document, calculate the difference between the average price and the price marked in the corresponding equipment list to obtain the price difference of each type of equipment, and then calculate the ratio of the average price of each type of equipment to obtain the horizontal price deviation of each type of equipment. The dynamic price volatility and horizontal price deviation of each type of equipment are compared with the preset dynamic price volatility threshold and horizontal price deviation threshold. Equipment with dynamic price volatility greater than the dynamic price volatility threshold or horizontal price deviation greater than the horizontal price deviation threshold is recorded as abnormal price quotation equipment.

8. The method for identifying and analyzing abnormal bidding behavior in power tendering as described in claim 7, characterized in that: The specific analysis method for the quotation anomaly scoring is as follows: The number of devices with abnormal pricing is counted, and then the ratio of this number to the total number of devices of all types is used to calculate the percentage of devices with abnormal risk. The average price anomaly is calculated by averaging the price anomalies of each abnormally quoted device. The average of the percentage of abnormal risk devices and the average price abnormality is used to calculate the price abnormality score.

9. The method for identifying and analyzing abnormal bidding behavior in power tendering as described in claim 1, characterized in that: The specific method for outputting the comprehensive risk index corresponding to each bidding document is as follows: The analysis yielded parameter deviation scores, operating condition mismatch scores, and price anomaly scores for each bidding document. The parameter deviation score, operating condition mismatch score, and price anomaly score are weighted and integrated to obtain the comprehensive risk index for each bidding document.

10. The method for identifying and analyzing abnormal bidding behavior in power tendering as described in claim 9, characterized in that: The specific analytical method for determining whether there is any abnormal bidding behavior in each bidding document is as follows: The comprehensive risk index of each bidding document is compared with the preset comprehensive risk index threshold. At the same time, the corresponding parameter deviation score, working condition mismatch score and price abnormality score are compared with the preset parameter deviation score threshold, working condition mismatch score threshold and price abnormality score threshold, respectively. If the comprehensive risk index is greater than the corresponding threshold or any score is greater than the corresponding threshold, the corresponding bid document is determined to have abnormal bidding behavior; otherwise, the corresponding bid document is determined not to have abnormal bidding behavior.

Images