Small and medium-sized enterprise single project accounting method based on industry financial minimum data unit

Abstract

The invention relates to the technical field of project accounting, in particular to a small and medium-sized enterprise single-project accounting method based on a business financial minimum data unit, which comprises the following steps of: matching collection and fulfillment nodes, distributing account affiliation, marking change of difference between income and expenditure directions, generating a cash flow recursion path, and uniformly writing in details of a total classification account to obtain accounting chain structure data. According to the method, time matching and amount identification of project collection information are realized by mapping contract node amounts and actual collection data in sequence, a node amount association chain with clear affiliation is constructed in combination with a mode of matching receivable nodes with account amounts step by step, and a linkage judgment mechanism of voucher numbers and revenue and expenditure identification is matched; according to the method, the income and expenditure directions of each stage of the project are distinguished, the fund change trajectory is reflected by sequence number difference operation, the directions of positive and negative items of cash flow are clear, and the response speed and accuracy of single-project accounting in dynamic fund tracking, cost control and accounting concentration are effectively improved.

Description

Technical Field

[0001] This invention relates to the field of project accounting technology, and in particular to a single-project accounting method for small and medium-sized enterprises based on the smallest data unit of business and finance. Background Technology

[0002] The field of project accounting technology involves the systematic collection, classification, and accounting of costs, revenues, expenses, and related financial information of various projects in enterprise operations. This includes project cost collection, expense allocation rule setting, accounting subject setting, financial data centralization, and fund flow records during the project execution cycle. Among these, the single-project accounting method for SMEs refers to the process by which SMEs, in order to achieve project fund management and cost control during the operation of a single project, manually establish accounting records or use the project auxiliary accounting function in general financial software to classify data and collect costs at the accounting subject level through project coding, voucher identification, and journal entry association, and complete the project accounting process based on the accounting entries within a fixed accounting period.

[0003] In the current project accounting process, the accounting activities mainly rely on the accounting entries within a fixed accounting period. In actual operation, it is difficult to dynamically match the actual collection progress with the contract receivable milestones. Especially when the collection milestones are received earlier or later, manual marking and voucher entry are prone to aggregation errors. There is a lack of automatic classification ability in the direction of income and expenditure. During the project, the flow of funds does not have a logical recursive chain, and the accounting attribution path is ambiguous. This leads to a loss of focus in cost control and an inability to trace changes in project funds. For example, if a project's intermediate collection is split into multiple payments, the traditional method can only aggregate by total amount, lacking a detailed allocation logic, which ultimately affects the timeliness and accounting accuracy of financial statements. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies by proposing a single-project accounting algorithm for small and medium-sized enterprises based on the smallest data unit of business finance.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a single-project accounting algorithm for SMEs based on the smallest data unit of business and finance, comprising the following steps: S1: Obtain the project number, contract performance obligation amount, and contract payment node sequence corresponding to the contract payment node, match them one by one with the actual payment date, mark and arrange the receivable amount and the received data in sequence, and generate a node sequence amount mapping record. S2: Based on the node sequence amount mapping record, extract the amount and time of each payment received, group them by project number, match the amount receivable according to the contract node sequence mark, compare the amount received from the first node in the interval, perform full amount allocation, and generate a node recursive attribution sequence number set. S3: Combining the recursive sequence number set of the nodes with the corresponding payment records, sort each record according to the payment time, combine and judge the voucher number and income / expenditure identifier, distinguish the direction of income and expenditure, and generate the project data unit income / expenditure mark record. S4: Based on the income and expenditure mark records of the project data unit, perform difference calculation on the amounts of adjacent sequential numbers and use it as the change benchmark, set the amount direction according to the current income and expenditure mark, mark the positive and negative items of the cash flow statement of the corresponding path, and generate the node recursive difference path.

[0006] As a further embodiment of the present invention, the node sequence amount mapping record includes a project number mapping set, an accounts receivable sorting list, and an arrival time pairing table; the project node allocation table includes an arrival amount range allocation structure, an attribution node identifier list, and a contract node mapping table; the project data unit income and expenditure mark record includes a voucher number sequence, a loan direction label, and an income and expenditure classification label; and the node recursive difference path includes an amount change difference set, a cash flow direction identifier sequence, and a sequence number path chain.

[0007] As a further aspect of the present invention, the specific steps for obtaining the node sequence amount mapping record are as follows: S111: Based on the contract payment node information and its corresponding project number, combined with the contract performance obligation amount data, the order of contract payment nodes is matched with the actual payment date, and the contract performance obligation amount is indexed and matched with the project number according to the node order to obtain the node performance amount order mapping table. S112: Based on the node performance amount sequence mapping table, combined with the obtained node sequence and the actual payment date, each payment node under each project number is matched with the received data and the amount receivable data, and the performance deviation degree of each node is identified and processed to obtain the amount deviation intensity under the project number dimension. S113: Based on the amount offset intensity under the project number dimension, perform structured annotation processing in conjunction with the project number, construct a two-way mapping between amount and node number according to the order of payment nodes, and generate a node order amount mapping record.

[0008] As a further aspect of the present invention, the specific steps for obtaining the node recursive attribution sequence number set are as follows: S211: Based on the node sequence amount mapping record, extract the amount and time of each receipt, group them by project number field, sort the receipt records within the same project number in ascending order by receipt time, add them to the account sequence number identifier, associate the node sequence of the receipt amount and the amount receivable, and obtain the receipt node matching index set. S212: Based on the payment node matching index set, extract the amount receivable under each node in the order of contract nodes, and establish the node sequential cumulative amount. Compare the amount received with the amount receivable of the current node in turn. If the amount received is less than or equal to the amount receivable of the current node, terminate the judgment process and write the attribution identifier. If it is greater, advance the difference sequentially to the next node. Calculate and obtain the difference intensity value of each payment amount recursively assigned to the node, and merge it with the project number and payment sequence number to generate a node attribution judgment result set. S213: Based on the node attribution judgment result set, according to the position range occupied by each payment amount in the node sequence, integrate all node numbers in the attribution judgment into a continuous number set, establish the node path corresponding to each payment amount, group and write it into the identifier set according to the project number and payment sequence number, and generate a node recursive attribution sequence number set.

[0009] As a further aspect of the present invention, the specific steps for obtaining the income and expenditure marker records of the project data unit are as follows: S311: Based on the recursive sequence number set of the nodes, extract the project number, loan direction identifier, voucher number and arrival time data contained in each record, sort them in ascending order according to the arrival time, establish the arrival time sequence index number, pair the voucher number with the loan direction identifier one-to-one, and obtain the voucher receipt and payment direction mark index set. S312: Based on the voucher number, arrival time and debit / credit direction identification information recorded in the voucher receipt and payment direction marking index set, construct a combination structure sorted by arrival order for all receipt records under the same item number, extract the correspondence between the voucher number and debit / credit direction in each combination item, identify the difference in the sorting position of debit / credit direction in the voucher sequence, and uniformly collect and identify the difference relationship under the item number dimension to obtain receipt and payment offset strength data; S313: Based on the difference identifier records collected in the income and expenditure offset intensity data, merge the voucher numbers within each project number, extract voucher groups with similar sorting offset values ​​as similar event clusters, perform classification operations according to the frequency ratio of occurrence in the lending direction, determine the income category and expenditure category, and generate income and expenditure mark records for project data units.

[0010] As a further aspect of the present invention, the specific steps for obtaining the node recursive difference path are as follows: S411: Based on the income and expenditure mark records of the project data unit, extract the sequence number, amount received and income and expenditure identifier marked in each record, classify and sort them according to the project number dimension, identify the increase or decrease between the previous sequence amount and the next sequence amount, aggregate all the difference pairs under the composition relationship, and obtain the set of adjacent sequence amount difference values. S412: Based on the adjacent sequential amount difference set, extract the sequential number and the marked income and expenditure identifier corresponding to each record, and perform grouping processing operation according to the directionality of each record to construct the pairing relationship between the sequential number and the direction identifier, and record the difference between the node number and the direction status to obtain the sequential node direction offset record. S413: Based on the path information structure in the sequential node direction offset record, mark all sequential numbers within each path, determine the direction mark status, pair all node identifier values ​​with sequential numbers, perform node hierarchical merging and number reorganization to obtain the node recursive difference path.

[0011] As a further aspect of the present invention, the method further includes: S5: Based on the path number, difference amount direction, node ownership sequence number and project number in the node recursive difference path, perform chain combination sorting according to the node ownership sequence number, and use the sequence number field as the connection index, write the project number as the anchor point into the general ledger details to generate single project dynamic accounting chain structure data. The single-project dynamic accounting chain structure data includes a path number combination table, a difference amount attribution structure, and a general ledger detail index set.

[0012] As a further aspect of the present invention, the specific steps for obtaining the single-project dynamic accounting chain structure data are as follows: S511: Based on the node recursive difference path, the path numbers are grouped and sorted in ascending order according to the node belonging number field within each group of paths to construct a continuous node sequence structure. Based on the project number as the classification basis, the sorting structure of all paths is summarized and integrated to obtain the belonging node order structure. S512: Based on the ordered structure of the attribution nodes, add an ordered number identifier to the position of each node number within the path, classify the change status of the amount direction in each node path, identify the directional continuity within the path, construct the association of the node structure under each path, summarize the node connection information of all paths, and generate an ordered connection record. S513: Based on the node path connection information summarized in the sequential connection record, the path structure under each project number dimension is reconstructed in detail. Nodes are sorted and combined according to the sequential number field. The loan / borrowing flag corresponding to the amount direction field is identified and the structure is transformed. Detailed assembly and writing are performed using the project number as the primary classification key to obtain the dynamic accounting chain structure data for a single project.

[0013] Compared with the prior art, the advantages and positive effects of the present invention are as follows: In this invention, by mapping contract node amounts to actual receipt data sequentially, the time matching and amount identification of project receipt information are achieved. Combined with the method of matching the received amount to the receivable node level by level, a clear node amount association chain is constructed. With the linkage judgment mechanism of voucher number and income and expenditure identification, the income and expenditure direction of each stage of the project is accurately distinguished. The difference calculation of sequential number reflects the trajectory of fund changes and clarifies the positive and negative cash flow. This achieves a financial information processing structure with fine data granularity, closed-loop path logic, clear income and expenditure direction, and complete accounting chain, effectively improving the response speed and accuracy of single project accounting in dynamic fund tracking, cost control, and accounting aggregation. Attached Figure Description

[0014] Figure 1 This is a flowchart of the main steps of the present invention; Figure 2 This is a flowchart of the process for obtaining the node sequence amount mapping record in this invention; Figure 3 This is a flowchart of the process for obtaining the node recursive attribution sequence number set in this invention; Figure 4 This is a flowchart illustrating the process of obtaining income and expenditure marker records for the data unit of this invention. Figure 5 This is a flowchart of the node recursive difference path acquisition process of the present invention; Figure 6 This is a flowchart of the data acquisition process for the single-project dynamic accounting chain structure of the present invention. Detailed Implementation

[0015] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0016] In the description of this invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, in the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0017] Please see Figure 1 The single-project accounting method for SMEs based on the smallest data unit of business and finance includes the following steps: S1: Obtain the project number, contract performance obligation amount (in accordance with the five-step revenue recognition model), and contract payment node sequence corresponding to the contract payment node, match them one by one with the actual payment date (execute the revenue recognition timing standard), use the project number as the index identifier, mark and arrange the receivable amount and the received data in sequence, and generate a node sequence amount mapping record; S2: Based on the node sequence amount mapping record, extract the amount and time of each payment received, group them by project number, and match the receivable amount according to the contract node sequence mark. Compare the amount received from the first node in the interval. If the amount is less than or equal to the receivable amount of the current node, set the node identifier and write it into the project node allocation table. If it is greater, proceed to the next node for recursive judgment according to the difference until the amount is completely allocated, and generate a set of node recursive ownership sequence numbers. S3: Combine the receipt records corresponding to the node recursive sequence number set, extract the project number, loan direction identifier (in accordance with financial business message standards), voucher number and receipt time of each record, sort each record according to the receipt time, set the sequence number to combine and judge the voucher number and income and expenditure identifier, distinguish the income and expenditure direction to form income and expenditure classification, and generate the project data unit income and expenditure mark record. S4: Based on the income and expenditure mark records of the project data unit, perform difference calculation on the amounts of adjacent sequential numbers and use it as the basis for change. Set the amount direction according to the current income and expenditure mark, mark the positive and negative items of the cash flow statement of the corresponding path (refer to the cash flow statement preparation standard), and generate the node recursive difference path. S5: Based on the path number, difference amount direction, node ownership number and project number in the node recursive difference path, perform chain combination sorting according to the node ownership number, and use the sequence number field as the connection index, write the data into the general ledger details (conforming to the camt.053 message format) with the project number as the anchor point, and generate single-project dynamic accounting chain structure data.

[0018] The node sequence amount mapping record includes a project number mapping set, an accounts receivable sorting list, and an arrival time pairing table. The project node allocation table includes an arrival amount range allocation structure, an attribution node identifier list, and a contract node mapping table. The project data unit income and expenditure mark record includes a voucher number sequence, a debit / credit direction label, and an income and expenditure classification label. The node recursive difference path includes an amount change difference set, a cash flow direction identifier sequence, and a sequence number path chain. The single project dynamic accounting chain structure data includes a path number combination table, a difference amount attribution structure, and a general ledger detail index set.

[0019] Please see Figure 2 The specific steps of S1 are as follows: S111: Based on the contract payment node information and its corresponding project number, combined with the contract performance obligation amount data, the order of contract payment nodes is matched with the actual payment date, and the contract performance obligation amount is indexed and matched with the project number according to the node order to obtain the node performance amount order mapping table. Based on the contract payment node information and its corresponding project number, the node fields in the original contract text must first be broken down item by item. The node identifier, performance obligation amount, and node sequence number field corresponding to each project number must be extracted. These fields can be extracted using the structured data fields recorded in the system. For example, contract A contains project number P001, with node information records N01 to N04, performance amounts of 50,000, 70,000, 60,000, and 80,000 respectively, and node sequences of 1, 2, 3, and 4. A structured parsing tool automatically extracts fields from the form format, mapping the node sequence field to the project number field one-to-one, thus obtaining the node performance amount structure for each project. For each node, the performance obligation amount field is combined with the arrival time field in the actual payment data to perform a binding operation between the node order and the arrival time. Differences in node order across different projects are transformed into a standardized sorting method for node numbers. The order is determined by comparing the time interval data between N01 and N02 in project P001 to see if it matches the system record. For example, if the arrival time of N01 is May 1, 2025, and the arrival time of N02 is July 1, 2025, the order corresponds, and the original sequence number is maintained; otherwise, the numbers need to be reordered. This process is repeated for multiple project numbers to complete the binding of node order and performance obligation amount. Finally, a corresponding data table is generated for each project number, showing the relationship between its performance obligation amount and node order, as shown below: Table 1 Mapping Table of Project Node Performance Obligations As shown in Table 1, each row of data will specifically reflect the sequential mapping information of a certain project node and its performance obligation amount. The generation process of this table provides an accurate index basis for further judging the performance completion status of each node and obtaining the sequential mapping table of node performance amount.

[0020] S112: Based on the node performance amount sequence mapping table, combined with the obtained node sequence and the actual payment date, each payment node under each project number is matched with the received data and the amount receivable data, and the performance deviation degree of each node is identified and processed to obtain the amount deviation intensity under the project number dimension. Based on the node performance amount sequence mapping table, it is necessary to match the performance obligation amount and the received amount under each node one by one. The system automatically compares the "receipt time" and "receipt amount" fields in the receipt data table, determines the stage according to the node sequence, extracts the received amount data, and processes it to match it with the node performance obligation amount to determine whether there is an offset at the amount level. For example, under project P001, the receivable amount of node N02 is 70,000 yuan, and the received amount record is 68,000 yuan, with a difference of 2,000 yuan. After extracting the amount offset for each node in this way, the amount offset value of each node is marked and stored in the data list. Then, combined with the time interval between the receipt time and the performance time, it is identified whether the current offset has a lagging distribution or an early offset. Based on this, the location of the offset and the offset intensity level are marked, and the offset intensity is divided into 0-3000 as low offset, 3000-6 000 represents a medium offset, and values ​​greater than 6000 represent a high offset. The offset level for each node is calculated. For example, in project P001, node N04 has an outstanding balance of 80,000 yuan, but only 74,000 yuan has been received, resulting in an offset of 6,000 yuan, which is classified as a medium offset. This process is repeated to iterate through all node information, outputting the offset levels for all nodes under each project number. Further, the offset levels of multiple nodes under the same project number are aggregated and integrated into a unified offset level array. For example, the array for project P001 is [low, medium, low, medium], and for project P002 it is [low, low]. The array information is then structurally analyzed to identify whether there is a continuous offset trend within the overall project. Based on the total cumulative offset value of the nodes and the median offset level, corresponding level classifications are assigned, and offset intensity level labels are assigned. Finally, the amount of offset intensity under the project number dimension is obtained.

[0021] S113: Based on the amount offset intensity under the project number dimension, perform structured annotation processing in conjunction with the project number, construct a two-way mapping between amount and node number according to the order of payment nodes, and generate node order amount mapping records; After determining the amount offset strength based on the project number dimension, it is necessary to reconstruct the bidirectional mapping structure between node order and amount based on the previously generated node order amount data. First, each node number and its sequence identifier are read sequentially using the project number as the dimension. Then, combined with the offset strength level information, it is merged with the node number to form a structured annotation format. The node order numbers are aggregated, and the node numbers under the same project are arranged in order. The correspondence between the node numbers and the corresponding performance amount is output in list form, further identifying the offset level field. For example, the node order for project P001 is [N01, N02, N...]. Given [N03, N04], with performance amounts of [50000, 70000, 60000, 80000] and offset levels of [low, medium, low, medium], the output structure is a dictionary format {P001: [(N01, 50000, low), (N02, 70000, medium), (N03, 60000, low), (N04, 80000, medium)]}. By writing the above structure into a data table and establishing a project number index in the system, a fast mapping call between project node information and amount values ​​and offset levels is realized, ultimately generating a node sequence amount mapping record.

[0022] Please see Figure 3 The specific steps of S2 are as follows: S211: Based on the node sequence amount mapping record, extract the amount and time of each receipt, group them by project number field, sort the receipt records within the same project number in ascending order by receipt time, add them to the account sequence number identifier, associate the node sequence of the receipt amount and the amount receivable, and obtain the receipt node matching index set. Based on the node-sequenced amount mapping records, the first step is to extract the amount received for each transaction and its corresponding arrival time field. Then, group the data according to the project number. Within each project group, sort the data in ascending order based on the value of the arrival time field, constructing an arrival sequence number field for the same group. In the structured system, this sequence number is expressed using array indices or numbers to represent the order of arrival. For example, in project P001, three transactions with arrival times of 20 days, 45 days, and 70 days are recorded, resulting in the sorted sequence [N01, N02, N03], labeled with arrival sequence numbers 1, 2, and 3 respectively. 3. Subsequently, these serial numbers, together with the original project number field, are used as a composite primary key to establish an index relationship between the amount received and the project number. Then, the index result is mapped one-to-one with the node order and the amount receivable, forming a sequential matching structure between the node's receivable amount and the actual amount received. This structure reflects the basic sequential position of the amount received within the node structure and serves as the matching input for subsequent attribution determination. Based on this mapping structure, the logical relationship and relative offset between the amount received and the node's amount structure can be quickly located. To support clear data display, the actual test data is listed below: Table 2 Project Payment Node Data Table As shown in Table 2, the arrival time, amount receivable and node order of different nodes are clearly correlated. This table provides a basic data index structure for determining the amount attribution and calculating the offset, and finally obtains the matching index set of the arrival node.

[0023] S212: Based on the matching index set of receiving nodes, extract the receivable amount under each node in the order of contract nodes, and establish a node-sequential cumulative amount. Compare the received amount with the current node's receivable amount sequentially. If the received amount is less than or equal to the current node's receivable amount, terminate the judgment process and write the attribution identifier. If it is greater, proceed sequentially to the next node, using the formula: ; The algorithm calculates the difference strength value for each received amount, recursively assigns nodes, and combines this value with the project number and the receipt sequence number to generate a node attribution determination result set. Indicates the first The intensity of the difference in allocation for each recursive node formed by the amount received, in yuan. Indicates the first The amount is deduced to the first The remaining unmatched amount at each node, in yuan. Indicates the first The amount receivable at each node, in yuan. This represents the absolute value of the amount offset at that node. Indicates the first The amount received and the first The time difference between the receivable times of each node, in days. Indicates the first The time span of contract nodes between a node and its predecessor, in days; After matching the index set based on the payment arrival nodes, the sequence of accounts receivable amounts needs to be extracted according to the contract node order structure. This sequence is then compared level by level with each payment arrival amount. If the current payment arrival amount is less than or equal to the current node's accounts receivable amount, the amount belongs to that node. If it is greater, the difference is pushed to the next node, and the same comparison process is repeated. This comparison process requires binding the difference between the payment arrival time field and the accounts receivable time field. Simultaneously, extract the receivable interval time between each node and the previous node. The proportion of these two factors is used in subsequent derivations to determine whether there is any inter-period shift in the amount, and the calculation is performed using a formula.

[0024] Taking the second payment for project P001 as an example, the amount received is 50,000 yuan, the corresponding node is N02, the amount receivable is 40,000 yuan, the arrival time is 45 days, and the node's due date is 50 days. Therefore: ; Substituting into the formula, the allocation term for this node is: ; If the funds are still not fully credited, proceed to node N03. Assume the remaining amount is 10,000 yuan, the amount due is 40,000 yuan, the arrival time is 70 days, and the due date is 75 days. : ; Summing all node items yields: ; thus, This reflects the overall recursive path strength of the funds, which will then be used for attribution structure calculations to ultimately generate a set of node attribution judgment results.

[0025] The strength value of the distribution difference between the recursive nodes of the received amount is a metric used to measure the overall deviation of a received amount from the contract node sequence and the pre-set structure of the receivable amount and the receivable time during the actual distribution process. This value comprehensively considers the difference between the amount and the receivable amount at each node during the distribution process, as well as the time interval between the actual receipt time of the amount and the receivable time of each distribution node. Specifically, it is expressed as the cumulative amount formed by weighting the amount deviation at each node in the node distribution path by the corresponding time misalignment factor. Therefore, the larger the strength value, the more nodes the funds have crossed in the distribution process or the greater the time deviation from the original nodes, reflecting the instability and non-sequentiality of the distribution in the recursive path. It is an important quantitative indicator for assessing the rationality of the amount distribution and the tightness of the node matching structure.

[0026] The formula characterizes the recursive distribution strength of received amounts in the node sequence based on a combination of amount offset and time misalignment. Indicates the first The funds were allocated to the first At each node, the absolute value of the difference between the remaining amount and the amount receivable is used to quantify the matching error or offset strength at the monetary level. This is used to measure whether the offset is accompanied by intertemporal behavior in the time structure, that is, the interval between the arrival time and the expected receivable time at the node. With the node's own time span By establishing a proportional relationship and taking the square root, the influence of linear growth on numerical amplification is compressed, allowing the impact of time mismatch to participate in the overall evaluation with non-linear weights. The multiplication operation couples the intensity of monetary offset with the intensity of time mismatch, thereby reflecting the overall offset intensity. Finally, by summing the offset product terms of all recursive nodes, the cumulative offset degree on the entire recursive path is reflected. Therefore, the product form of "monetary error × inter-period factor" is adopted, and the overall allocation difference is obtained by accumulating at the node level.

[0027] S213: Based on the node attribution judgment result set, according to the position range occupied by each payment amount in the node sequence, integrate all node numbers in the attribution judgment into a set of consecutive numbers, establish the node path corresponding to each payment amount, group and write it into the identifier set according to the project number and payment sequence number, and generate a set of node recursive attribution sequence numbers. Based on the node attribution judgment result set, a path set of nodes actually occupied by each received amount needs to be constructed. The system reads the receipt sequence number and attribution judgment result under each project number, and marks the allocation path in array form by identifying all node numbers between the starting node position and the ending node position. For example, the first received amount under project P001 is 30,000 yuan, which belongs only to node N01. The amount receivable is 40,000 yuan, which is insufficient to be recursively pushed to subsequent nodes, so its attribution path is [N01]. For the second fund of 50,000 yuan, it belongs to node N02 and overflows to N03, according to the previous text. The calculation process shows that its inter-period offset strength is significant, and the system labels its belonging path as [N02, N03]. The third fund of 40,000 yuan is completely consistent with the amount of node N03, belonging to a single node, and the path is [N03]. All the above belonging paths are written into the mapping structure to form a node recursive structured set. The final output is in the following labeling format: The path set corresponding to project P001 is: Serial number 1 → [N01]; Serial number 2 → [N02, N03]; Serial number 3 → [N03]; Summarize the above path numbering structure, establish a project dimension attribution mapping dictionary and record the index relationship, and generate a node recursive attribution sequence set.

[0028] Please see Figure 4 The specific steps of S3 are as follows: S311: Based on the recursive sequence number set of the nodes, extract the project number, loan direction identifier, voucher number and arrival time data contained in each record, sort them in ascending order according to the arrival time, establish the arrival time sequence index number, pair the voucher number with the loan direction identifier one-to-one, and obtain the voucher receipt and payment direction mark index set. Based on the recursive sequence number set of the nodes corresponding to the received records, the project number, debit / credit direction identifier, voucher number, and arrival time are extracted from each record. First, the records are sorted in ascending order according to the arrival time field to construct the arrival order structure. For example, under project P001, vouchers V001 have an arrival time of 12 days and V002 has an arrival time of 18 days, so the sorting result is numbered 1 and 2. Under project P002, vouchers V004 have an arrival time of 20 days and V003 has an arrival time of 22 days, so the corresponding order numbers are 1 and 2. After sorting, the voucher number and debit / credit direction identifier fields are matched one by one to determine whether the debit / credit direction is "credit". Only records with the direction of "credit" are kept, and the rest are discarded, as shown in Table 3. Table 3. Voucher Record Data Table In Table 3, debit record V002 under P001 is removed, and only V001, V003, and V004 are retained. Record index numbers are constructed in sequence, and the result records are summarized to obtain the voucher receipt and payment direction mark index set.

[0029] S312: Based on the voucher number, arrival time and debit / credit direction identification information recorded in the voucher receipt and payment direction mark index set, construct a combination structure sorted by arrival order for all receipt records under the same item number, extract the correspondence between the voucher number and debit / credit direction in each combination item, identify the difference in the sorting position of debit / credit direction in the voucher sequence, and uniformly collect and identify the difference relationship under the item number dimension to obtain receipt and payment offset strength data. Based on the voucher receipt and payment direction marker index set, a sequential combination structure is constructed for all vouchers under the project number dimension. The arrival time and sequential number of each record are matched one-to-one with the debit and credit direction. The arrangement of the debit and credit direction in the arrival sequence is identified. Taking P002 as an example, voucher V004 has an arrival time of 20 days and a sequence number of 1, while v003 has an arrival time of 22 days and a sequence number of 2. Both vouchers have the debit and credit direction of "credit", but the arrival order is different, thus forming a sorting offset pair. The system records their time difference and sequence difference, and groups and aggregates each offset pair. If there is a case where the arrival time is earlier but the sorting is later or the order is reversed, the degree of offset is marked as a non-zero difference value. Under the project dimension, a set of sorting structure differences among multiple voucher pairs is formed, and their distribution is statistically analyzed to finally obtain the receipt and payment offset intensity data.

[0030] S313: Based on the difference identifier records collected in the income and expenditure offset intensity data, merge the voucher numbers within each project number, extract voucher groups with similar sorting offset values ​​as similar event clusters, perform classification operations according to the frequency ratio of occurrence in the lending direction, determine the income category and expenditure category, and generate income and expenditure mark records for project data units. Based on the income and expenditure offset intensity data, the voucher records within the project number dimension are grouped. The income and expenditure offset judgment benchmark is set as a time difference of no more than 5 days. If the arrival time difference of two vouchers is less than or equal to 5 and their sequential numbers are consecutive, they are judged to be the same event combination. For example, under P002, the arrival time difference of V004 and V003 is 2 days, and their sequential numbers are 1 and 2. Therefore, they meet the combination condition and are grouped into the same group. Then, the frequency of the debit / credit direction of "credit" in this combination is counted, and the recognition threshold is set to 60%. If the frequency of "credit" in the group is not less than 60%, it is classified as income; otherwise, it is classified as expenditure. This threshold is set based on the general structural proportion of income combination in income and expenditure data. After the combination direction is determined, the classification mark is written into the project data structure to form a complete record field, and finally the income and expenditure mark record of the project data unit is obtained.

[0031] Please see Figure 5 The specific steps of S4 are as follows: S411: Based on the income and expenditure mark records of project data units, extract the sequence number, amount received and income and expenditure identifier marked in each record, classify and sort them according to the project number dimension, identify the increase and decrease of the amount in the previous sequence and the amount in the next sequence, aggregate the difference pairs under all the constituent relationships, and obtain the set of difference values ​​of adjacent sequence amounts. Based on the project data unit income and expenditure marker records, the project number, sequence number, amount received, and income / expenditure identifier fields are extracted from the records. First, records within each project number are sorted in ascending order of sequence number. Then, it is determined whether the sequence numbers are consecutive. If the sequence difference is 1, the records are considered consecutive; otherwise, the combination is excluded and not included in the subsequent difference extraction process. For consecutively numbered record pairs, the amount received corresponding to the next number is compared with the amount received by the previous number to identify the trend of amount change and record the difference value. Simultaneously, the sequence number combination information and the project number to which each record pair belongs are retained. In project P007, the amounts received for sequence numbers 1, 2, and 3 are 10,000 yuan, 13,000 yuan, and 9,000 yuan respectively, with a difference of +3,000 yuan between numbers 1 and 2, and -4,000 yuan between numbers 2 and 3. In project P008, the amounts corresponding to sequence numbers 1 and 2 are 20,000 yuan and 16,000 yuan respectively, with a difference of -4,000 yuan. The combination of these difference pairs forms the following structure: Table 4. Receipt Records and Income / Expense Markings As shown in Table 4, the system identifies the monetary difference structure in adjacent sequential numbers, forms a structure set, and inputs it into the next logic module to obtain the adjacent sequential monetary difference set.

[0032] S412: Based on the set of adjacent sequential amount differences, extract the sequential number and the marked income and expenditure identifier corresponding to each record, and perform grouping processing operation according to the directionality of each record to construct the pairing relationship between the sequential number and the direction identifier, and record the difference between the node number and the direction status to obtain the sequential node direction offset record. Based on the set of adjacent sequential amount differences, the sequential number and corresponding income / expenditure identifier associated with each difference record are read one by one. According to the identifier content, "income" records are regarded as inflows and "expenditure" records are regarded as outflows. A node direction sequence is constructed under the project number dimension. Taking project P007 as an example, numbers 1 and 2 are identified as "income" and number 3 is identified as "expenditure". Then the direction from node 1 to 2 is positive and the direction from 2 to 3 is negative. In this path, the direction changes from continuous positive to negative, and the direction state changes once. The system records the sequential number of the direction change and the identifier state before and after the change. If the direction identifier does not change, the direction offset is zero. If there is a change, the offset identifier is one. Through the comparison structure of the entire path, the number of direction mutations in each path is counted and merged and collected. Finally, the set of changes in the direction structure is obtained, and the sequential node direction offset record is generated.

[0033] S413: Based on the path information structure in the sequential node direction offset record, mark all sequential numbers within each path, determine the direction mark status, pair all node identifier values ​​with sequential numbers, perform node hierarchical merging and number reorganization to obtain the node recursive difference path. Based on the direction status within the path identified by the sequential node direction offset record, the sorted node numbers in each path are labeled with directions. The node numbers and their corresponding direction identifiers are extracted. Within each project path, nodes with the direction identifier "receive" correspond to cash inflows, and nodes with the direction identifier "expenditure" correspond to cash outflows. The system structurally combines the node numbers and direction types to form a pairing structure between node numbers and cash flow directions. This pairing is then used to form a direction path sequence. If there is a direction change within the path, the node direction status is switched and marked in the node structure. If the direction identifiers remain consistent, the path direction status is a continuous structure. This structure serves as a reference item for the final cash flow compilation path and is written into the system structure field, ultimately generating the node recursive difference path.

[0034] Please see Figure 6 The specific steps of S5 are as follows: S511: Based on the node recursive difference path, the path numbers are grouped and sorted in ascending order according to the node belonging number field within each group to construct a continuous node sequence structure. The sorting structure of all paths is summarized and integrated based on the project number to obtain the belonging node order structure. Based on the four fields included in the node recursive difference path—path number, difference amount direction, node ownership number, and project number—the path number field is grouped. Then, under each path number, the node ownership number field is sorted in ascending order. The sorted node numbers and their corresponding amount direction information in the path are extracted one by one, and their original project number values ​​are synchronously collected and recorded. During the sorting process, it is necessary to determine whether the node ownership number is a consecutive integer. If there are skipped ownership numbers in a group, such as numbers 1, 3, and 4, they should be recorded as two independent sequences to avoid incorrect combinations of subsequent paths due to non-continuous chains. For each sorted path structure, it is divided into the corresponding project group according to the project number field. In actual operation, taking project P013 as an example, its path T01 has node ownership numbers 1, 2, and 3, and the amount direction is "positive, negative, positive". The node with the belonging number 1 has a positive monetary direction and a corresponding monetary value of 3000 yuan; the node with the belonging number 2 has a negative monetary direction and a monetary value of -1000 yuan; and the node with the belonging number 3 has a positive monetary direction and a monetary value of 2000 yuan. After aggregation, they form the path sequence structure T01→P013: (1, 3000, positive), (2, -1000, negative), (3, 2000, positive). At the same time, path T02 in project P014 contains the belonging numbers 1 and 2 of the nodes, with monetary directions of negative and positive respectively, forming the structure T02→P014: (1, -1500, negative), (2, 2500, positive). This structure is merged into the path group corresponding to project number P014. Under multiple project numbers, after constructing a complete node chain sort, the sorting information of all path numbers and the belonging node sequence number are combined and incorporated into the project dimension structure to obtain the belonging node sequence structure.

[0035] S512: Based on the sequential structure of the attribution nodes, an sequential number identifier is added to the position of each node number within the path, the change status of the amount direction in each node path is classified and processed, the directional continuity within the path is identified, the node structure under each path is associated and constructed, the node connection information of all paths is summarized, and a sequential connection record is generated. Based on the sequential structure of the attribution nodes, each node is assigned a unique sequential number as an index field for node connection. Simultaneously, the original monetary direction identifier of each node is extracted as a direction attribute record within each path, and a structural anchoring relationship is established in conjunction with the project number field. Based on this, the monetary direction status in each path is categorized and analyzed to determine whether there is a state transition in monetary direction between adjacent nodes. If the monetary direction of node i is positive and the direction of node i+1 is negative, it is counted as a direction switching event. The cumulative number of transitions in the path serves as the basic indicator of the degree of direction change. For example, if the directions of the three nodes in path T01 are positive, negative, and positive respectively, then the number of direction transitions is 2. The direction of path T02 is... Negative and positive values ​​result in one transition. If the direction of path T03 is continuously positive, the number of transitions is 0. The number of transitions is recorded in the path structure and combined with its sequential number. At the same time, the direction identifiers that appear most frequently in the direction combination are recorded as path tendency indicators. For example, if all 3 nodes in a path are negative, the path tendency is expenditure-oriented. If it is 2 positive and 1 negative, it is income-oriented. The direction tendency judgment interval is set to 60% or more of the nodes having the same identifier, which is considered to be the dominant direction. After performing node index field normalization, direction transition identification, and direction tendency classification operations on each path structure, a node sequence and direction connection state structure is constructed in all path dimensions, and finally the sequence connection record is obtained.

[0036] S513: Based on the node path connection information summarized in the sequential connection record, the path structure under each project number dimension is reconstructed in detail, the nodes are sorted and combined according to the sequential number field, the loan and borrowing flags corresponding to the amount direction field are identified and the structure is transformed, and the details are assembled and written according to the project number as the main classification key to obtain the dynamic accounting chain structure data of a single project. Based on the path direction structure information recorded in the sequential connection record, the sequence field and amount direction field of the corresponding node number for each path number are called, and combined in ascending order according to the node sequence field to form a linear structure record. Then, the project number field is used as the classification anchor field to integrate all path structure records under the same project number into a unified data structure. The field structure requirements in the CAMT.053 message format standard are called to map the field of each path record item to the corresponding detail field. For example, the sequence number is mapped to "EntryReference" in the message, the amount direction is mapped to "CreditDebitIndicator", and the amount value corresponds to "Amount". The path number is entered as "AdditionalEntryInfo". The project number is used to construct the "BatchBooking" field. After all fields are mapped within the structural standard, the field combination output is performed on each path structure. During the combination process, it is necessary to identify whether the field content meets the character length and field limit range. If it does not meet the requirements, the data format is converted. For example, if the amount field requires two decimal places, the amount "3000" needs to be converted to "3000.00". Then, all node fields in the path are combined into a message structure node chain according to the sequential number. Finally, the node chain structure under all project numbers is summarized to generate the single project dynamic accounting chain structure data.

[0037] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A single-project accounting algorithm for SMEs based on the smallest data unit of business and finance, characterized in that, Includes the following steps: S1: Obtain the project number, contract performance obligation amount, and contract payment node sequence corresponding to the contract payment node, match them one by one with the actual payment date, mark and arrange the receivable amount and the received data in sequence, and generate a node sequence amount mapping record. S2: Based on the node sequence amount mapping record, extract the amount and time of each payment received, group them by project number, match the amount receivable according to the contract node sequence mark, compare the amount received from the first node in the interval, perform full amount allocation, and generate a node recursive attribution sequence number set. S3: Combining the recursive sequence number set of the nodes with the corresponding payment records, sort each record according to the payment time, combine and judge the voucher number and income / expenditure identifier, distinguish the direction of income and expenditure, and generate the project data unit income / expenditure mark record. S4: Based on the income and expenditure mark records of the project data unit, perform difference calculation on the amounts of adjacent sequential numbers and use it as the change benchmark, set the amount direction according to the current income and expenditure mark, mark the positive and negative items of the cash flow statement of the corresponding path, and generate the node recursive difference path.

2. The single-project accounting algorithm for SMEs based on the smallest business and financial data unit according to claim 1, characterized in that, The node sequence amount mapping record includes a project number mapping set, an accounts receivable sorting list, and an arrival time pairing table. The project node allocation table includes an arrival amount range allocation structure, an attribution node identifier list, and a contract node mapping table. The project data unit income and expenditure mark record includes a voucher number sequence, a loan direction label, and an income and expenditure classification label. The node recursive difference path includes an amount change difference set, a cash flow direction identifier sequence, and a sequence number path chain.

3. The single-project accounting algorithm for SMEs based on the smallest business and financial data unit according to claim 1, characterized in that, The specific steps for obtaining the node sequential amount mapping record are as follows: S111: Based on the contract payment node information and its corresponding project number, combined with the contract performance obligation amount data, the order of contract payment nodes is matched with the actual payment date, and the contract performance obligation amount is indexed and matched with the project number according to the node order to obtain the node performance amount order mapping table. S112: Based on the node performance amount sequence mapping table, combined with the obtained node sequence and the actual payment date, each payment node under each project number is matched with the received data and the amount receivable data, and the performance deviation degree of each node is identified and processed to obtain the amount deviation intensity under the project number dimension. S113: Based on the amount offset intensity under the project number dimension, perform structured annotation processing in conjunction with the project number, construct a two-way mapping between amount and node number according to the order of payment nodes, and generate a node order amount mapping record.

4. The single-project accounting algorithm for SMEs based on the smallest business and financial data unit according to claim 1, characterized in that, The specific steps for obtaining the node recursive belonging sequence number set are as follows: S211: Based on the node sequence amount mapping record, extract the amount and time of each receipt, group them by project number field, sort the receipt records within the same project number in ascending order by receipt time, add them to the account sequence number identifier, associate the node sequence of the receipt amount and the amount receivable, and obtain the receipt node matching index set. S212: Based on the payment node matching index set, extract the amount receivable under each node in the order of contract nodes, and establish the node sequential cumulative amount. Compare the amount received with the amount receivable of the current node in turn. If the amount received is less than or equal to the amount receivable of the current node, terminate the judgment process and write the attribution identifier. If it is greater, advance the difference sequentially to the next node. Calculate and obtain the difference intensity value of each payment amount recursively assigned to the node, and merge it with the project number and payment sequence number to generate a node attribution judgment result set. S213: Based on the node attribution judgment result set, according to the position range occupied by each payment amount in the node sequence, integrate all node numbers in the attribution judgment into a continuous number set, establish the node path corresponding to each payment amount, group and write it into the identifier set according to the project number and payment sequence number, and generate a node recursive attribution sequence number set.

5. The single-project accounting algorithm for SMEs based on the smallest business and financial data unit according to claim 1, characterized in that, The specific steps for obtaining the income and expenditure tag records of the project data unit are as follows: S311: Based on the recursive sequence number set of the nodes, extract the project number, loan direction identifier, voucher number and arrival time data contained in each record, sort them in ascending order according to the arrival time, establish the arrival time sequence index number, pair the voucher number with the loan direction identifier one-to-one, and obtain the voucher receipt and payment direction mark index set. S312: Based on the voucher number, arrival time and debit / credit direction identification information recorded in the voucher receipt and payment direction marking index set, construct a combination structure sorted by arrival order for all receipt records under the same item number, extract the correspondence between the voucher number and debit / credit direction in each combination item, identify the difference in the sorting position of debit / credit direction in the voucher sequence, and uniformly collect and identify the difference relationship under the item number dimension to obtain receipt and payment offset strength data; S313: Based on the difference identifier records collected in the income and expenditure offset intensity data, merge the voucher numbers within each project number, extract voucher groups with similar sorting offset values ​​as similar event clusters, perform classification operations according to the frequency ratio of occurrence in the lending direction, determine the income category and expenditure category, and generate income and expenditure mark records for project data units.

6. The single-project accounting algorithm for SMEs based on the smallest business and financial data unit according to claim 1, characterized in that, The specific steps for obtaining the node recursive difference path are as follows: S411: Based on the income and expenditure mark records of the project data unit, extract the sequence number, amount received and income and expenditure identifier marked in each record, classify and sort them according to the project number dimension, identify the increase or decrease between the previous sequence amount and the next sequence amount, aggregate all the difference pairs under the composition relationship, and obtain the set of adjacent sequence amount difference values. S412: Based on the adjacent sequential amount difference set, extract the sequential number and the marked income and expenditure identifier corresponding to each record, and perform grouping processing operation according to the directionality of each record to construct the pairing relationship between the sequential number and the direction identifier, and record the difference between the node number and the direction status to obtain the sequential node direction offset record. S413: Based on the path information structure in the sequential node direction offset record, mark all sequential numbers within each path, determine the direction mark status, pair all node identifier values ​​with sequential numbers, perform node hierarchical merging and number reorganization to obtain the node recursive difference path.

7. The single-project accounting algorithm for SMEs based on the smallest business and financial data unit according to claim 1, characterized in that, The method further includes: S5: Based on the path number, difference amount direction, node ownership sequence number and project number in the node recursive difference path, perform chain combination sorting according to the node ownership sequence number, and use the sequence number field as the connection index, write the project number as the anchor point into the general ledger details to generate single project dynamic accounting chain structure data. The single-project dynamic accounting chain structure data includes a path number combination table, a difference amount attribution structure, and a general ledger detail index set.

8. The single-project accounting algorithm for SMEs based on the smallest business and financial data unit according to claim 7, characterized in that, The specific steps for obtaining the single-project dynamic accounting chain structure data are as follows: S511: Based on the node recursive difference path, the path numbers are grouped and sorted in ascending order according to the node belonging number field within each group of paths to construct a continuous node sequence structure. Based on the project number as the classification basis, the sorting structure of all paths is summarized and integrated to obtain the belonging node order structure. S512: Based on the ordered structure of the attribution nodes, add an ordered number identifier to the position of each node number within the path, classify the change status of the amount direction in each node path, identify the directional continuity within the path, construct the association of the node structure under each path, summarize the node connection information of all paths, and generate an ordered connection record. S513: Based on the node path connection information summarized in the sequential connection record, the path structure under each project number dimension is reconstructed in detail, the nodes are sorted and combined according to the sequential number field, the loan and borrowing flags corresponding to the amount direction field are identified and the structure is transformed, and the details are assembled and written according to the project number as the main classification key to obtain the single project dynamic accounting chain structure data.

Images