Whole-process digital auditing system for electric power engineering project

Abstract

The invention relates to a whole-process digital auditing system for an electric power engineering project. The whole-process digital auditing system comprises a data acquisition unit, a data storage unit, an auditing analysis unit, a process automation unit and a visualization unit, the data acquisition unit acquires financial data in real time through an electric power ERP system, and the financial data comprises budget, expenditure, cost and contract information; the data storage unit adopts a relational database for storing structured financial data; the auditing analysis unit is used for identifying abnormal transactions in the financial data by using a machine learning algorithm and discovering potential fraudulent behaviors or errors; the process automation unit is used for automatically executing repetitive data input and checking tasks by utilizing a robot process automation technology; and the visualization unit uses Power BI to establish a financial data visualization instrument board, and visually displays key information and audit discovery. The financial management and auditing capability of the electric power engineering project is effectively improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent auditing, and in particular to a full-process digital auditing system for electric power engineering projects. Background Art

[0002] In the wave of modern digital transformation, large infrastructure projects such as power engineering projects are facing increasingly complex management and compliance challenges. These projects usually involve a large amount of financial data, contracts and multiple parties involved. Accurate and timely management of this information is crucial to project success and legal compliance.

[0003] The existing audit system relies on a large amount of manual operations and paper documents to record and track financial transactions, which increases the possibility of human error and operational complexity. It relies more on rules and sample audits, making it difficult to detect more complex and hidden frauds. It lacks the support of intelligent tools and cannot fully utilize big data analysis to improve the depth and breadth of audit findings. Summary of the invention

[0004] In order to solve the above problems, the purpose of the present invention is to provide a full-process digital audit system for power engineering projects, which not only improves the financial management and audit capabilities of power engineering projects, but also realizes the efficiency and transparency of business processes.

[0005] To achieve the above object, the present invention adopts the following technical solutions:

[0006] A full-process digital audit system for power engineering projects, including a data acquisition unit, a data storage unit, an audit analysis unit, a process automation unit and a visualization unit;

[0007] The data acquisition unit acquires financial data in real time through the power ERP system, including budget, expenditure, cost and contract information;

[0008] The data storage unit pre-processes the data acquired by the data acquisition unit and uses a relational database (such as PostgreSQL) to store structured financial data;

[0009] The audit analysis unit uses a machine learning algorithm to identify abnormal transactions in financial data and discover potential fraud or errors;

[0010] The process automation unit uses robotic process automation technology to automatically perform repetitive data input and verification tasks, thereby improving efficiency and reducing human errors;

[0011] The visualization unit uses PowerBI to build a financial data visualization dashboard to intuitively display key information and audit findings.

[0012] Furthermore, preprocessing includes removing duplicate records from the dataset, making each transaction unique in the database, filling missing values ​​using the mean, and standardizing the data.

[0013] Furthermore, the machine learning algorithm uses LOF to identify abnormal transactions in financial data, as follows:

[0014] Determine the features analyzed by the LOF algorithm, including transaction amount and transaction frequency, and aggregate the feature information to form the calculated data matrix X;

[0015] For each data point p, find the k nearest neighbors, and the distance metric is Euclidean distance;

[0016] The reachable distance is defined as the k-reachable distance of data point p to data point o:

[0017] reach-distk(p,o)=max(k-dist(o),dist(p,o));

[0018] Among them, k-dist(o) is the distance from o to its kth nearest neighbor, and dist(p,o) is the Euclidean distance from point p to o;

[0019] Calculate the local reachability density, LRD k (p) is the local reachability density of point pp, which is the reciprocal mean of the reachability distances of its neighbors:

[0020]

[0021] Calculate the local anomaly factor of each point, which is the average of the ratio of its local reachability density to the reachability density of its neighboring points:

[0022]

[0023] Among them, if LOF k (p)>1, then point p is abnormal relative to its neighbors;

[0024] A reasonable LOF threshold is set according to business requirements. When the LOF value of a data point exceeds the threshold, it is marked as abnormal, and further business analysis is performed on the screened abnormal points to determine whether there is potential fraud or error.

[0025] Further, robotic process automation technology is as follows:

[0026] Identify highly repetitive, standardized, and rule-defined tasks in the process, including financial data import, verification, reconciliation, and report generation;

[0027] Record and analyze existing manual processes in detail to identify inputs and outputs, key steps, decision points, and exception handling solutions; use workflow tools to visualize the process;

[0028] Based on the RPA tool UiPath, confirm the tool's supported data input and output formats, script or interface recording functions, exception handling capabilities, and integration capabilities with ERP.

[0029] Design robot scripts based on process mapping and refine the execution of each step, including data input, clicks, data verification and exception handling;

[0030] Establish a robust exception handling mechanism for common errors or system failures, including retry, logging, and alarm functions;

[0031] Scripts automate data entry, extract data from documents using OCR technology or import data from ERP;

[0032] Automatically check budget and actual expenditure data, and verify data accuracy using preset rules such as formula consistency and range checks.

[0033] Furthermore, according to the process mapping, the robot script is designed to refine the execution operations of each step, including data input, click, data verification and exception handling, as follows:

[0034] Describe each step in detail, clarify the input, operation, output and its conditions, including business rules, decision conditions and data verification standards; Based on the RPA tool UiPath, batch read data from the ERP system and use OCR technology to process data in scanned documents or pictures;

[0035] Simulate manual data input by filling out forms, selecting drop-down options, etc.; define fields: including locating input boxes, identifying through XPath or UI element ID; filling in data, confirming that the input is accurate, identifying the operation location, completing user interface operations by simulating clicks and keyboard input, locating buttons or links through the properties of UI elements; ensuring that operations are correctly executed according to business processes;

[0036] Use the Try-Catch structure: define a standard exception handling process; log records: record in detail the input data, step location, and error information when an exception occurs; retry and downgrade processing: set automatic retry logic or forward manual processing; retry strategy: define the retry interval and number of times; downgrade path: when automatic processing is not possible, mark the task as requiring manual intervention.

[0037] Furthermore, PowerBI is used to build a financial data visualization dashboard, as follows:

[0038] Define data models in PowerBI, create data relationships and hierarchies, and ensure that the relationships between all related tables are correct to support joint analysis and data interaction

[0039] Determine the core purpose of the dashboard, identifying key metrics and information to be presented, including budget, expenditure, cost and contract information;

[0040] Use interactive features in PowerBI (such as slicers and filters) to allow users to dynamically view data of different dimensions; implement drill-down functions to allow users to drill down from the overview layer to detailed data and achieve hierarchical information browsing;

[0041] Use PowerBI's chart types to display data distribution and trends.

[0042] The present invention has the following beneficial effects:

[0043] 1. The present invention not only improves the financial management and auditing capabilities of power engineering projects, but also makes business processes more efficient and transparent;

[0044] 2. The present invention uses machine learning algorithms to detect anomalies in financial data, which can automatically discover potential fraud or errors, reduce the risk of financial errors, improve audit efficiency and accuracy, and use robotic process automation (RPA) technology to automate repetitive tasks, significantly improve efficiency, reduce errors in manual operations, and release human resources for higher-value analysis and decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0046] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments:

[0047] refer to Figure 1 ,In this implementation, a full process digital audit system for a power engineering project is provided, including a data acquisition unit, a data storage unit, an audit and analysis unit, a process automation unit, and a visualization unit;

[0048] The data acquisition unit acquires financial data in real time through the power ERP system, including budget, expenditure, cost and contract information;

[0049] The data storage unit pre-processes the data acquired by the data acquisition unit and uses a relational database (such as PostgreSQL) to store structured financial data;

[0050] The audit analysis unit uses a machine learning algorithm to identify abnormal transactions in financial data and discover potential fraud or errors;

[0051] The process automation unit uses robotic process automation technology to automatically perform repetitive data input and verification tasks, thereby improving efficiency and reducing human errors;

[0052] The visualization unit uses PowerBI to build a financial data visualization dashboard to intuitively display key information and audit findings.

[0053] In this embodiment, the preprocessing includes deleting duplicate records in the data set, making each transaction unique in the database, using the mean to fill in missing values, and standardizing the data.

[0054] In this embodiment, the machine learning algorithm uses LOF to identify abnormal transactions in financial data, as follows:

[0055] Determine the features analyzed by the LOF algorithm, including transaction amount and transaction frequency, and aggregate the feature information to form the calculated data matrix X;

[0056] For each data point p, find the k nearest neighbors, and the distance metric is Euclidean distance;

[0057] The reachable distance is defined as the k-reachable distance of data point p to data point o:

[0058] reach-distk(p,o)=max(k-dist(o),dist(p,o));

[0059] Among them, k-dist(o) is the distance from o to its kth nearest neighbor, and dist(p,o) is the Euclidean distance from point p to o;

[0060] Calculate the local reachability density, LRD k (p) is the local reachability density of point pp, which is the reciprocal mean of the reachability distances of its neighbors:

[0061]

[0062] Calculate the local anomaly factor of each point, which is the average of the ratio of its local reachability density to the reachability density of its neighboring points:

[0063]

[0064] Among them, if LOF k (p)>1, then point p is abnormal relative to its neighbors;

[0065] A reasonable LOF threshold is set according to business requirements. When the LOF value of a data point exceeds the threshold, it is marked as abnormal, and further business analysis is performed on the screened abnormal points to determine whether there is potential fraud or error.

[0066] In this embodiment, the robotic process automation technology is as follows:

[0067] Identify highly repetitive, standardized, and rule-defined tasks in the process, including financial data import, verification, reconciliation, and report generation;

[0068] Record and analyze existing manual processes in detail to identify inputs and outputs, key steps, decision points, and exception handling solutions; use workflow tools to visualize the process;

[0069] Based on the RPA tool UiPath, confirm the tool's supported data input and output formats, script or interface recording functions, exception handling capabilities, and integration capabilities with ERP.

[0070] Design robot scripts based on process mapping and refine the execution of each step, including data input, clicks, data verification and exception handling;

[0071] Establish a robust exception handling mechanism for common errors or system failures, including retry, logging, and alarm functions;

[0072] Scripts automate data entry, extract data from documents using OCR technology or import data from ERP;

[0073] Automatically check budget and actual expenditure data, and verify data accuracy using preset rules such as formula consistency and range checks.

[0074] In this embodiment, according to the process mapping, the robot script is designed to refine the execution operation of each step, including data input, click, data verification and exception handling, as follows:

[0075] Describe each step in detail, clarify the input, operation, output and its conditions, including business rules, decision conditions and data verification standards; Based on the RPA tool UiPath, batch read data from the ERP system and use OCR technology to process data in scanned documents or pictures;

[0076] Simulate manual data input by filling out forms, selecting drop-down options, etc.; define fields: including locating input boxes, identifying through XPath or UI element ID; filling in data, confirming that the input is accurate, identifying the operation location, completing user interface operations by simulating clicks and keyboard input, locating buttons or links through the properties of UI elements; ensuring that operations are correctly executed according to business processes;

[0077] Use the Try-Catch structure: define a standard exception handling process; log records: record in detail the input data, step location, and error information when an exception occurs; retry and downgrade processing: set automatic retry logic or forward manual processing; retry strategy: define the retry interval and number of times; downgrade path: when automatic processing is not possible, mark the task as requiring manual intervention.

[0078] In this embodiment, PowerBI is used to build a financial data visualization dashboard, as follows:

[0079] Define data models in PowerBI, create data relationships and hierarchies, and ensure that the relationships between all related tables are correct to support joint analysis and data interaction

[0080] Determine the core purpose of the dashboard, identifying key metrics and information to be presented, including budget, expenditure, cost and contract information;

[0081] Use interactive features in PowerBI (such as slicers and filters) to allow users to dynamically view data of different dimensions; implement drill-down functions to allow users to drill down from the overview layer to detailed data and achieve hierarchical information browsing;

[0082] Use PowerBI's chart types to display data distribution and trends.

[0083] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0084] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0085] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0086] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0087] The above is only a preferred embodiment of the present invention, and does not limit the present invention in other forms. Any technician familiar with the profession may use the above disclosed technical content to change or modify it into an equivalent embodiment with equivalent changes. However, any simple modification, equivalent change and modification made to the above embodiment according to the technical essence of the present invention without departing from the technical solution of the present invention still belongs to the protection scope of the technical solution of the present invention.

Claims

1. A full-process digital audit system for power engineering projects, characterized in that: It includes data acquisition unit, data storage unit, audit analysis unit, process automation unit and visualization unit; The data acquisition unit acquires financial data in real time through the power ERP system, including budget, expenditure, cost and contract information; The data storage unit pre-processes the data acquired by the data acquisition unit and uses a relational database to store structured financial data; The audit analysis unit uses a machine learning algorithm to identify abnormal transactions in financial data and discover potential fraud or errors; The process automation unit uses robotic process automation technology to automatically perform repetitive data input and verification tasks, thereby improving efficiency and reducing human errors; The visualization unit uses PowerBI to build a financial data visualization dashboard to intuitively display key information and audit findings.

2. The whole process digital audit system of a power engineering project according to claim 1 is characterized in that: The preprocessing includes deleting duplicate records in the data set, making each transaction unique in the database, using the mean to fill in missing values, and standardizing the data.

3. The whole process digital audit system of a power engineering project according to claim 1 is characterized in that: The machine learning algorithm uses LOF to identify abnormal transactions in financial data, as follows: Determine the features analyzed by the LOF algorithm, including transaction amount and transaction frequency, and aggregate the feature information to form the calculated data matrix X; For each data point p, find the k nearest neighbors, and the distance metric is Euclidean distance; The reachable distance is defined as the k-reachable distance of data point p to data point o: reach-distk(p,o)=max(k-dist(o),dist(p,o)); Among them, k-dist(o) is the distance from o to its kth nearest neighbor, and dist(p,o) is the Euclidean distance from point p to o; Calculate the local reachability density, LRD k (p) is the local reachability density of point pp, which is the reciprocal mean of the reachability distances of its neighbors: Calculate the local anomaly factor of each point, which is the average of the ratio of its local reachability density to the reachability density of its neighboring points: Among them, if LOF k (p)>1, then point p is abnormal relative to its neighbors; A reasonable LOF threshold is set according to business requirements. When the LOF value of a data point exceeds the threshold, it is marked as abnormal, and further business analysis is performed on the screened abnormal points to determine whether there is potential fraud or error.

4. The whole process digital audit system of a power engineering project according to claim 1 is characterized in that: The robotic process automation technology is specifically as follows: Identify highly repetitive, standardized, and rule-defined tasks in the process, including financial data import, verification, reconciliation, and report generation; Record and analyze existing manual processes in detail to identify inputs and outputs, key steps, decision points, and exception handling plans; Use workflow tools to visualize the process; Based on the RPA tool UiPath, confirm the data input and output formats, script or interface recording functions, exception handling capabilities, and integration capabilities with ERP supported by the tool; Design robot scripts based on process mapping and refine the execution of each step, including data input, clicks, data verification and exception handling; Establish a robust exception handling mechanism for common errors or system failures, including retry, logging, and alarm functions; Scripts automate data entry, extract data from documents using OCR technology or import data from ERP; Automatically check budget and actual expenditure data, and verify data accuracy using preset rules such as formula consistency and range checks.

5. A full-process digital audit system for electric power engineering projects according to claim 4, characterized in that: According to the process mapping, the robot script is designed to refine the execution operation of each step, including data input, click, data verification and exception handling, as follows: Describe each step in detail, clarify the input, operation, output and its conditions, including business rules, decision conditions and data verification standards; Based on the RPA tool UiPath, batch read data from the ERP system and use OCR technology to process data in scanned documents or pictures; Simulate manual data input by filling out forms, selecting drop-down options, etc.; define fields: including locating input boxes, identifying through XPath or UI element ID; filling in data, confirming that the input is accurate, identifying the operation location, completing user interface operations by simulating clicks and keyboard input, locating buttons or links through the properties of UI elements; ensuring that operations are correctly executed according to business processes; Use the Try-Catch structure: define a standard exception handling process; log records: record in detail the input data, step location, and error information when an exception occurs; retry and downgrade processing: set automatic retry logic or forward manual processing; Retry strategy: define the retry interval and number of times; Degradation path: when it cannot be processed automatically, mark the task as requiring manual intervention.

6. The whole process digital audit system of a power engineering project according to claim 1 is characterized in that: The use of PowerBI to build a financial data visualization dashboard is as follows: Define data models in PowerBI, create data relationships and hierarchies, and ensure that the relationships between all related tables are correct to support joint analysis and data interaction Determine the core purpose of the dashboard, identifying key metrics and information to be presented, including budget, expenditure, cost and contract information; Use the interactive function in PowerBI to allow users to dynamically view data of different dimensions; implement the drill-down function to allow users to drill down from the overview layer to detailed data and realize hierarchical information browsing; Use Power BI chart types to display data distribution and trends.

Images