An AI-based automatic settlement method for fuel contracts

Abstract

The present application relates to the field of fuel contract settlement technology, and in particular to an AI-based automatic settlement method for fuel contracts. It includes: generating training set data based on historical fuel contract settlement parameters, and establishing a settlement agent based on the training set data; judging whether all pending contracts are in a settlement cycle, and establishing a pending contract sequence A based on the judgment result; the settlement agent generates a settlement sequence based on the pending contract sequence A, and sets the settlement parameters of each pending contract based on the settlement sequence; by classifying fuel contracts, the settlement agent uses different settlement strategies to process different contract types, which improves the settlement efficiency of the contract on the one hand, and timely warns of the risks that may arise in the settlement of each contract on the other hand, thereby reducing the possible default losses after settlement.

Description

Technical Field

[0001] The present application relates to the technical field of fuel contract settlement, and in particular to an AI-based automatic settlement method for fuel contracts. Background Art

[0002] At present, many thermal power generation enterprises have not yet established a scientific and comprehensive fuel control system, lack effective supervision means, and have not built an effective integrated information platform for fuel control. There are "information islands" in the relevant business links of fuel management such as planning, procurement, scheduling, acceptance, settlement, coal yard, blending, and analysis, and the information of each link is not effectively organized and interconnected. In these business links, a large amount of manual operation, recording, tabulation, calculation, analysis and other related work is required, which is labor-intensive and redundant. Coal quality, settlement and economic activity analysis are still at the level of manual work, which is time-consuming and labor-intensive, and consumes a lot of human resources. As a result, there are many problems in fuel management, especially in the acceptance, settlement, blending and other processes.

[0003] At present, when settling fuel contracts, data from documents such as quality and chemical analysis are manually extracted, and judgments are made manually based on the contract terms, and calculation sheets are filled out. Framework contract terms and multiple batch deliveries for a single contract increase the difficulty of preparing settlement sheets. At the same time, different types of contracts may face different risks when they are settled, and the current settlement technology is unable to analyze and promptly warn of risks in the contract performance process. Summary of the invention

[0004] The purpose of this application is: to solve the above-mentioned technical problems, this application provides an AI-based automatic settlement method for fuel contracts, aiming to improve the settlement efficiency of fuel contracts, and at the same time provide timely warnings for possible settlement risks of fuel contracts to reduce corporate losses.

[0005] In some embodiments of the present application, historical fuel contract parameters and AI technology are used to construct a settlement intelligent entity, which can intelligently extract the settlement terms in each fuel contract, establish a corresponding settlement model, complete the autonomous settlement of the fuel contract, and generate the corresponding settlement statement, thereby improving settlement efficiency and reducing labor costs.

[0006] In some embodiments of the present application, fuel contracts are classified and the settlement agent adopts different settlement strategies to handle different contract types. On the one hand, the settlement efficiency of the contract is improved, and on the other hand, timely warnings are given to the risks that may arise during the settlement of each contract, thereby reducing the possible default losses that may occur after the settlement.

[0007] In some embodiments of the present application, a method for automatic settlement of fuel contracts based on AI is provided, comprising:

[0008] Generate training set data based on historical fuel contract settlement parameters, and establish a settlement agent based on the training set data;

[0009] Determine whether all pending contracts are in the settlement cycle, and create a pending contracts sequence A based on the determination result;

[0010] The settlement agent generates a settlement sequence based on the sequence A of contracts to be settled, and sets the settlement parameters of each contract to be settled according to the settlement sequence;

[0011] Where A=(a1,a2…an),n is the number of pending contracts in the settlement cycle at the current time node, and ai is the i-th pending contract at the current time node.

[0012] In some embodiments of the present application, when generating a pending settlement sequence, the process includes:

[0013] According to the series A of contracts to be settled, set the contracts to be settled ai as the contracts to be evaluated in sequence;

[0014] Generate a reference evaluation value sequence Bi of the contract to be evaluated, Bi=(b1,b2…bm), where bi is the reference evaluation value of the i-th evaluation index of the contract to be evaluated; m is the number of evaluation indexes;

[0015] Generate the settlement evaluation value c of the contract to be evaluated according to the reference evaluation value sequence B;

[0016] ; Among them, αi is the weight coefficient of the i-th evaluation index;

[0017] Generate a settlement evaluation value sequence C, C = (c1, c2...cn), where ci is the settlement evaluation value of the i-th contract to be settled;

[0018] Generate the first-level settlement order according to the settlement evaluation value sequence C.

[0019] In some embodiments of the present application, when setting the settlement parameters of each contract to be settled, it includes:

[0020] The settlement agent selects a contract to be settled according to the first settlement order, and sets the selected contract to be settled as a target contract to be settled;

[0021] Get the contract type of the target contract to be settled;

[0022] If the target contract to be settled is a first-level settlement contract, a first-level settlement instruction is generated;

[0023] If the target contract to be settled is a secondary settlement contract, a secondary settlement instruction is generated;

[0024] If the target contract to be settled is a level 3 settlement contract, a level 3 settlement instruction is generated;

[0025] Generate settlement parameters for the target contract to be settled based on the settlement instruction.

[0026] In some embodiments of the present application, the primary settlement instruction includes:

[0027] The settlement agent extracts the settlement clause parameters of the target contract to be settled;

[0028] Generate a computational thinking chain based on the settlement terms parameters and establish a primary settlement model for the target contract to be settled;

[0029] Obtain the performance parameters of the target contract to be settled and generate document data based on the performance parameters;

[0030] Generate the first-level settlement document of the target contract to be settled based on the first-level settlement model and document data.

[0031] In some embodiments of the present application, the secondary settlement instruction includes:

[0032] The settlement agent extracts the identity label and expected transaction volume q of the performance party in the target contract to be settled;

[0033] Obtain the historical performance parameters of the performance object according to the identity tag of the performance object;

[0034] According to the historical performance parameters, the delivery evaluation value series F of the performance object is established, F=(f1,f2…fβ), where fi is the delivery evaluation value of the i-th fuel contract of the performance object, and β is the number of fuel contracts signed with the performance object;

[0035] Generate security transaction volume q1 based on historical performance parameters;

[0036] If q>q1, ​​generate the volume difference Δq;

[0037] Generate a performance evaluation value d for the performance object, and determine whether to generate a secondary settlement document based on the performance evaluation value d.

[0038] In some embodiments of the present application, the generating of the performance evaluation value d of the performance object includes:

[0039] d=e1*[ (fi*gi)]+e2*[ (fi-Δf) 2 ]+e3*Y(Δq);

[0040] Wherein, gi is the influencing factor of the i-th delivery evaluation value, e1 is the preset first fixed coefficient, e2 is the preset second fixed coefficient, e3 is the preset third fixed coefficient, Δf is the average value of the delivery evaluation value, and Y(x) is the default risk function.

[0041] In some embodiments of the present application, when determining whether to generate a secondary settlement statement, it includes:

[0042] The settlement agent extracts the settlement clause parameters of the target contract to be settled;

[0043] Establish a secondary settlement model for the target contract to be settled;

[0044] Preset the first performance evaluation value threshold D1 and the second performance evaluation value threshold D2;

[0045] When d < D1;

[0046] The settlement agent generates a secondary settlement statement according to the expected trading volume and the secondary settlement model;

[0047] When D1 ≤ d < D2;

[0048] The settlement agent generates a secondary settlement plan for the target contract to be settled;

[0049] Generate a secondary settlement statement according to the secondary settlement plan and the secondary settlement model;

[0050] When d ≥ D2, generate a secondary warning instruction and stop settlement.

[0051] In some embodiments of the present application, the tertiary settlement instruction includes:

[0052] The settlement agent extracts the performance clause parameters of the target contract to be settled;

[0053] According to the performance clause parameters, establish a sub-settlement cycle sequence T of the target contract to be settled, T = (t1, t2... t n1 ), where t i is the i-th sub-settlement cycle of the target contract to be settled; n1 is the number of sub-settlement cycles;

[0054] Select the sub-settlement cycle t n2 where the current time node is located, and 1 < n2 < n1;

[0055] Generate the first execution risk evaluation value H1 according to the performance parameters of all sub-settlement cycles before the sub-settlement cycle tn2;

[0056] Generate the second execution risk evaluation value H2 according to the performance progress of the target contract to be settled;

[0057] Generate a decision evaluation value k according to the first execution risk evaluation value H1 and the second execution risk evaluation value H2;

[0058] k=θ*(e4*H1+e5*H2),wherein, e4 is the preset fourth weight coefficient, e5 is the preset fifth weight coefficient, and θ is the correction coefficient set according to the performance object parameters of the target contract to be settled;

[0059] Determine whether to generate a third-level settlement statement based on the decision evaluation value k.

[0060] In some embodiments of the present application, the determination of whether to generate a third-level settlement document includes:

[0061] The settlement agent extracts the settlement clause parameters of the target contract to be settled;

[0062] Establish a three-level settlement model for target contracts to be settled;

[0063] Preset decision evaluation value threshold K1;

[0064] When k>K1, the settlement agent sub-settlement period t n2 The performance parameters of the contract are calculated and a three-level settlement document is generated according to the three-level settlement model;

[0065] When k≤K1, a third-level warning instruction is generated and settlement is stopped.

[0066] Compared with the prior art, the AI-based automatic settlement method for fuel contracts in the embodiment of the present application has the following beneficial effects:

[0067] By utilizing historical fuel contract parameters and AI technology, a settlement intelligent entity is constructed, which can intelligently extract the settlement terms in each fuel contract, establish a corresponding settlement model, complete the autonomous settlement of the fuel contract, and generate the corresponding settlement slip, thereby improving settlement efficiency and reducing labor costs.

[0068] By classifying fuel contracts, the settlement agent adopts different settlement strategies to handle different contract types. On the one hand, it improves the settlement efficiency of the contract, and on the other hand, it promptly warns of the risks that may arise during the settlement of each contract, thereby reducing the possible default losses after settlement. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 It is a flow chart of an AI-based automatic settlement method for fuel contracts in a preferred embodiment of the present application. DETAILED DESCRIPTION

[0070] The specific implementation methods of the present application are further described in detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present application but are not intended to limit the scope of the present application.

[0071] In the description of the present application, it should be understood that the terms "center", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present application.

[0072] The terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, unless otherwise specified, "plurality" means two or more.

[0073] In the description of this application, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0074] like Figure 1 As shown, a preferred embodiment of the present application embodiment is an AI-based automatic settlement method for fuel contracts, including:

[0075] S101: Generate training set data according to historical fuel contract settlement parameters, and establish a settlement agent according to the training set data;

[0076] S102: Determine whether all pending contracts are in a settlement cycle, and create a pending contract sequence A based on the determination result;

[0077] S103: The settlement agent generates a settlement sequence according to the sequence A of contracts to be settled, and sets settlement parameters of each contract to be settled according to the settlement sequence;

[0078] Where A=(a1,a2…an),n is the number of pending contracts in the settlement cycle at the current time node, and ai is the i-th pending contract at the current time node.

[0079] Specifically, its settlement intelligence entity is built based on AI technology and big data models. It can autonomously identify contract terms, derive settlement formulas based on settlement terms, and generate settlement data based on actual document information. It can automatically check the settlement data, verify error-prone points, and finally generate corresponding settlement orders to improve overall settlement efficiency.

[0080] Specifically, all pending contracts in pending contracts sequence A are within the settlement cycle.

[0081] Specifically, when generating the order to be settled, it includes:

[0082] According to the series A of contracts to be settled, set the contracts to be settled ai as the contracts to be evaluated in sequence;

[0083] Generate a reference evaluation value sequence Bi of the contract to be evaluated, Bi=(b1,b2…bm), where bi is the reference evaluation value of the i-th evaluation index of the contract to be evaluated; m is the number of evaluation indexes;

[0084] Generate the settlement evaluation value c of the contract to be evaluated according to the reference evaluation value sequence B;

[0085] ; Among them, αi is the weight coefficient of the i-th evaluation index;

[0086] Generate a settlement evaluation value sequence C, C = (c1, c2...cn), where ci is the settlement evaluation value of the i-th contract to be settled;

[0087] Generate the first-level settlement order according to the settlement evaluation value sequence C.

[0088] Specifically, the evaluation indicators include, but are not limited to, the time when the contract subject has been delivered, the contract subject quantity, the level of the performance object and the transaction amount. The larger the settlement evaluation value, the higher the settlement priority of the corresponding contract to be settled.

[0089] Specifically, by setting a first-level settlement order, each contract to be settled is sorted and settled in sequence according to their priority, thereby improving the overall settlement efficiency.

[0090] In a preferred embodiment of the present application, when setting the settlement parameters of each contract to be settled, it includes:

[0091] The settlement agent selects a contract to be settled according to the first settlement order, and sets the selected contract to be settled as a target contract to be settled;

[0092] Get the contract type of the target contract to be settled;

[0093] If the target contract to be settled is a first-level settlement contract, a first-level settlement instruction is generated;

[0094] If the target contract to be settled is a secondary settlement contract, a secondary settlement instruction is generated;

[0095] If the target contract to be settled is a level 3 settlement contract, a level 3 settlement instruction is generated;

[0096] Generate settlement parameters for the target contract to be settled based on the settlement instruction.

[0097] Specifically, a first-level settlement contract refers to a contract in which fuel is delivered first and payment is made later. When settling such a contract, there is no subsequent risk of fuel delivery, so settlement can be carried out in the usual way.

[0098] Specifically, a secondary settlement contract refers to a contract in which payment is made first and fuel is delivered later. Since the fuel is delivered later in this type of contract, it may cause losses to the company if the fuel is not delivered in time. Therefore, there is a fuel delivery risk when settling this type of contract.

[0099] Specifically, the third type of settlement contract refers to a contract for fuel delivery in multiple times. Since such contracts have a long delivery cycle and there are multiple settlement nodes, there are delivery risks at some settlement nodes.

[0100] Specifically, the first-level settlement instructions include:

[0101] The settlement agent extracts the settlement clause parameters of the target contract to be settled;

[0102] Generate a computational thinking chain based on the settlement terms parameters and establish a primary settlement model for the target contract to be settled;

[0103] Obtain the performance parameters of the target contract to be settled and generate document data based on the performance parameters;

[0104] Generate the first-level settlement document of the target contract to be settled based on the first-level settlement model and document data.

[0105] Specifically, the settlement agent extracts the settlement terms parameters, establishes the corresponding first-level settlement model, and generates document data for each location based on the recorded performance parameters, such as obtaining the fuel delivery quantity at the coal yard, the fuel delivery quality at the procurement part, and the calculated price at the financial entrance. It then generates corresponding settlement data based on the document data and the first-level settlement model, automatically checks the settlement data, and generates a first-level settlement form after checking the error-prone points. This improves settlement efficiency and settlement accuracy, and solves the problems of low efficiency and low accuracy of manual settlement.

[0106] In a preferred embodiment of the present application, the secondary settlement instruction includes:

[0107] The settlement agent extracts the identity label and expected transaction volume q of the performance party in the target contract to be settled;

[0108] Obtain the historical performance parameters of the performance object according to the identity tag of the performance object;

[0109] According to the historical performance parameters, the delivery evaluation value series F of the performance object is established, F=(f1,f2…fβ), where fi is the delivery evaluation value of the i-th fuel contract of the performance object, and β is the number of fuel contracts signed with the performance object;

[0110] Generate security transaction volume q1 based on historical performance parameters;

[0111] If q>q1, ​​generate the volume difference Δq;

[0112] Generate a performance evaluation value d for the performance object, and determine whether to generate a secondary settlement document based on the performance evaluation value d.

[0113] Specifically, the larger the delivery evaluation value, the worse the performance quality of the corresponding contract at the time of performance. The delivery evaluation value can be set based on multiple indicators such as the delivery quality at the time of performance and whether there is a delay in delivery.

[0114] Specifically, the safe transaction volume q1 can be set according to historical data. The safe transaction volume q1 means that when the delivery volume of the performance object is q1, the delivery quality is optimal and there will be no problems such as delayed delivery and substandard fuel quality. That is, there will be no breach of contract.

[0115] Specifically, when generating the performance evaluation value d, it includes:

[0116] d=e1*[ (fi*gi)]+e2*[ (fi-Δf) 2 ]+e3*Y(Δq);

[0117] Among them, gi is the influencing factor of the i-th delivery evaluation value, e1 is the preset first fixed coefficient, e2 is the preset second fixed coefficient, e3 is the preset third fixed coefficient, Δf is the average value of the delivery evaluation value, and Y(x) is the default risk function.

[0118] Specifically, Y(x) is the default risk coefficient with the trading volume difference as the independent variable. The larger the trading volume difference, the greater the corresponding default risk probability.

[0119] Specifically, by presetting the first fixed coefficient, the second fixed coefficient and the third fixed coefficient, [ (fi*gi)],[ (fi-Δf) 2and Y(Δq) are in the same reference system, and the larger the actual values of the three, the larger the corresponding performance evaluation value. The larger the performance evaluation value, the greater the possibility of default risk in the current contract to be performed.

[0120] Specifically, when determining whether to generate a secondary settlement statement, it includes:

[0121] The settlement agent extracts the settlement clause parameters of the target contract to be settled;

[0122] Establish a secondary settlement model for the target contract to be settled;

[0123] Preset the first performance evaluation value threshold D1 and the second performance evaluation value threshold D2;

[0124] When d < D1;

[0125] The settlement agent generates a secondary settlement statement according to the expected trading volume and the secondary settlement model;

[0126] When D1 ≤ d < D2;

[0127] The settlement agent generates a secondary settlement plan for the target contract to be settled;

[0128] Generate a secondary settlement statement according to the secondary settlement plan and the secondary settlement model;

[0129] When d ≥ D2, generate a secondary warning instruction and stop settlement.

[0130] Specifically, the secondary settlement plan is optimized according to relevant parameters to obtain the safe settlement volume at the current time node, set the settlement time node of the remaining settlement volume, and generate sub-settlement statements respectively. Finally, a secondary settlement statement is generated according to all sub-settlement statements. On the premise of ensuring settlement efficiency, reduce the losses caused by possible contract defaults. And timely warn of possible default risks by generating secondary warning instructions.

[0131] In the preferred embodiment of this application, the tertiary settlement instruction includes:

[0132] The settlement agent extracts the performance clause parameters of the target contract to be settled;

[0133] According to the performance clause parameters, establish a sub-settlement cycle sequence T of the target contract to be settled, T = (t1, t2... t n1 ), where ti is the i-th sub-settlement cycle of the target contract to be settled; n1 is the number of sub-settlement cycles;

[0134] Select the sub-settlement cycle t n2 where the current time node is located, and 1 < n2 < n1;

[0135] According to the sub-settlement period t n2 The performance parameters of all previous sub-settlement cycles generate the first execution risk assessment value H1;

[0136] Generate a second execution risk assessment value H2 based on the performance progress of the target contract to be settled;

[0137] Generate a decision evaluation value k according to the first execution risk evaluation value H1 and the second execution risk evaluation value H2;

[0138] k=θ*(e4*H1+e5*H2),wherein, e4 is the preset fourth weight coefficient, e5 is the preset fifth weight coefficient, and θ is the correction coefficient set according to the performance object parameters of the target contract to be settled;

[0139] Determine whether to generate a third-level settlement statement based on the decision evaluation value k.

[0140] Specifically, the first execution risk assessment value and the second execution risk assessment value have the same value range, and the larger the first execution risk assessment value, the higher the previous performance quality, and the larger the second execution risk assessment value, the faster the current performance speed.

[0141] Specifically, θ can be set according to the historical performance parameters of the performance object. The higher the historical performance quality, the higher the corresponding θ value.

[0142] Specifically, the larger the decision evaluation value, the lower the possibility of default in subsequent delivery.

[0143] Specifically, when determining whether to generate a third-level settlement document, it includes:

[0144] The settlement agent extracts the settlement clause parameters of the target contract to be settled;

[0145] Establish a three-level settlement model for target contracts to be settled;

[0146] Preset decision evaluation value threshold K1;

[0147] When k>K1, the settlement agent sub-settlement period t n2 The performance parameters of the contract are calculated and a three-level settlement document is generated according to the three-level settlement model;

[0148] When k≤K1, a third-level warning instruction is generated and settlement is stopped.

[0149] Specifically, when the decision evaluation value is lower than the preset threshold, it means that the current contract to be settled cannot complete subsequent delivery and settlement needs to be stopped, thereby reducing the company's losses.

[0150] According to the first concept of the present application, a settlement intelligent entity is constructed by utilizing historical fuel contract parameters and AI technology, which can intelligently extract the settlement terms in each fuel contract, establish a corresponding settlement model, complete the autonomous settlement of the fuel contract, and generate the corresponding settlement statement, thereby improving settlement efficiency and reducing labor costs.

[0151] According to the second concept of the present application, by classifying fuel contracts, the settlement intelligent body adopts different settlement strategies to handle different contract types. On the one hand, it improves the settlement efficiency of the contract, and on the other hand, it timely warns of the risks that may arise in the settlement of each contract, thereby reducing the possible default losses after settlement.

[0152] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and substitutions can be made without departing from the technical principles of the present application. These improvements and substitutions should also be regarded as the scope of protection of the present application.

Claims

1. An AI-based automatic settlement method for fuel contracts, characterized in that: include: Generate training set data based on historical fuel contract settlement parameters, and establish a settlement agent based on the training set data; Determine whether all pending contracts are in the settlement cycle, and create a pending contracts sequence A based on the determination result; The settlement agent generates a settlement sequence based on the sequence A of contracts to be settled, and sets the settlement parameters of each contract to be settled according to the settlement sequence; Where A=(a1,a2…an),n is the number of pending contracts in the settlement cycle at the current time node, and ai is the i-th pending contract at the current time node; When generating the order to be settled, it includes: According to the series A of contracts to be settled, set the contracts to be settled ai as the contracts to be evaluated in sequence; Generate a reference evaluation value sequence Bi of the contract to be evaluated, Bi=(b1,b2…bm), where bi is the reference evaluation value of the i-th evaluation indicator of the contract to be evaluated; m is the number of evaluation indicators; Generate the settlement evaluation value c of the contract to be evaluated according to the reference evaluation value sequence B; ; Among them, αi is the weight coefficient of the i-th evaluation index; Generate a settlement evaluation value sequence C, C = (c1, c2...cn), where ci is the settlement evaluation value of the i-th contract to be settled; Generate the first-level settlement order according to the settlement evaluation value sequence C; When setting the settlement parameters for each contract to be settled, include: The settlement agent selects a contract to be settled according to the first settlement order, and sets the selected contract to be settled as a target contract to be settled; Get the contract type of the target contract to be settled; If the target contract to be settled is a first-level settlement contract, a first-level settlement instruction is generated; If the target contract to be settled is a secondary settlement contract, a secondary settlement instruction is generated; If the target contract to be settled is a level 3 settlement contract, a level 3 settlement instruction is generated; Generate settlement parameters of the target contract to be settled according to the settlement instruction; The primary settlement instructions include: The settlement agent extracts the settlement clause parameters of the target contract to be settled; Generate a computational thinking chain based on the settlement terms parameters and establish a primary settlement model for the target contract to be settled; Obtain the performance parameters of the target contract to be settled and generate document data based on the performance parameters; Generate the first-level settlement document of the target contract to be settled based on the first-level settlement model and document data; The secondary settlement instructions include: The settlement agent extracts the identity label and expected transaction volume q of the performance party in the target contract to be settled; Obtain the historical performance parameters of the performance object according to the identity tag of the performance object; According to the historical performance parameters, the delivery evaluation value series F of the performance object is established, F=(f1,f2…fβ), where fi is the delivery evaluation value of the i-th fuel contract of the performance object, and β is the number of fuel contracts signed with the performance object; Generate security transaction volume q1 based on historical performance parameters; If q>q1, ​​generate the volume difference Δq; Generate a performance evaluation value d of the performance object, and determine whether to generate a secondary settlement document based on the performance evaluation value d; The generating of the performance evaluation value d of the performance object includes: d=e1*[ (fi*gi)]+e2*[ (fi-Δf) 2 ]+e3*Y(Δq); Wherein, gi is the influencing factor of the i-th delivery evaluation value, e1 is the preset first fixed coefficient, e2 is the preset second fixed coefficient, e3 is the preset third fixed coefficient, Δf is the average value of the delivery evaluation value, and Y(x) is the default risk function; When determining whether to generate a secondary settlement statement, it includes: The settlement agent extracts the settlement clause parameters of the target contract to be settled; Establish a secondary settlement model for the target contract to be settled; Preset the first performance evaluation value threshold D1 and the second performance evaluation value threshold D2; When d < D1; The settlement agent generates a secondary settlement statement based on the expected trading volume and the secondary settlement model; When D1 ≤ d < D2; The settlement agent generates a secondary settlement plan for the target contract to be settled; Generate a secondary settlement statement according to the secondary settlement plan and the secondary settlement model; When d ≥ D2, generate a secondary warning instruction and stop settlement; The said tertiary settlement instruction includes: The settlement agent extracts the performance clause parameters of the target contract to be settled; According to the performance clause parameters, the sub-settlement period sequence T of the target contract to be settled is established, T= (t1, t2…t n1 ), where ti is the i-th sub-settlement period of the target contract to be settled; n1 is the number of sub-settlement periods; Select the sub-settlement period t of the current time node n2 , and 1 <n2<n1; According to the sub-settlement period t n2 The performance parameters of all previous sub-settlement cycles generate the first execution risk assessment value H1; Generate a second execution risk evaluation value H2 according to the performance progress of the target contract to be settled; Generate a decision evaluation value k according to the first execution risk evaluation value H1 and the second execution risk evaluation value H2; k = θ * (e4 * H1 + e5 * H2), where e4 is a preset fourth weight coefficient, e5 is a preset fifth weight coefficient, and θ is a correction coefficient set according to the performance object parameters of the target contract to be settled; Judge whether to generate a tertiary settlement statement according to the decision evaluation value k; When determining whether to generate a tertiary settlement statement, it includes: The settlement agent extracts the settlement clause parameters of the target contract to be settled; Establish a tertiary settlement model for the target contract to be settled; Preset a decision evaluation value threshold K1; When k>K1, the settlement agent sub-settlement period t n2 The performance parameters of the contract are calculated and a three-level settlement document is generated according to the three-level settlement model; When k ≤ K1, generate a tertiary warning instruction and stop settlement.

Images