CN116308276A - Logic knowledge system for aircraft maintenance - Google Patents

Abstract

A logical knowledge system model for aircraft maintenance is disclosed, using answer set programming (Answer Set programming) to build a knowledge base and system. The knowledge base effectively describes the maintenance main body, maintenance resources, maintenance guidance and quality confirmation, and can effectively query maintenance guidance and maintenance quality confirmation criteria and methods by matching with maintenance tasks programmed and written by using the answer set. According to the result given by the system, the engineer can better study the relation between faults, and further improve the efficiency of the maintenance task of the aircraft.

Description

Logic knowledge system for aircraft maintenance

Technical Field

The invention belongs to the field of civil aviation engineering, and particularly relates to a logic knowledge system which is programmed by an answer set and is used for maintaining an aircraft.

Background

The maintenance of the aircraft is important to the stable and safe operation of the aircraft, and the airlines and airlines including boeing and air passengers in all countries pay great attention to the maintenance. With the continuous advent of new technologies and new technologies, the increasing complexity and diversity of aircraft structures has become increasingly difficult to maintain for an aircraft, and the early manufacture and maintenance during operation of an aircraft has become a complex area involving multiple disciplines and multiple branches, so there is a need to digitize and program these maintenance procedures and ideas. Taking the wave sound of a leader in the digitizing process as an example, a wave sound company develops a series of digital auxiliary technologies for reducing the development difficulty of new products and ensuring the performance and reliability of the new products, and digitizes the processing and assembly process planning of each part. Within airlines, digital assistance technology is focused on maintaining process control, optimizing inventory, optimizing maintenance costs, and the like. There is a lack of digitization on the market for specific maintenance methods and specific maintenance procedures.

The traditional maintenance method and the fault type are single, most of the maintenance work of the aircraft firstly needs maintenance personnel to make a maintenance plan and check, and then can carry out the maintenance work by applying for maintenance to the upper level, obtaining approval and signing files. Maintenance of parts on board an aircraft is typically performed by looking at operational data and manually determining if maintenance is required. The workers also perform repair by turning over the formed maintenance manual. Such a process consumes a lot of manpower and material resources and requires an increase in the speed of repair and maintenance by using modern means. Moreover, the existing aircraft has multiple fault types, complex relationship among parts and various fault reasons, and the parts needing to be maintained and the specific process of maintenance are difficult to judge by simple manual work. We therefore propose to use answer set programming (Answer Set programming) to construct a knowledge representation and reasoning system that provides corresponding maintenance guidelines according to different demand situations.

The maintenance process knowledge has the characteristic of dynamic property, and the maintenance knowledge has various forms. If a database system is used, unified assumption needs to be made on the data in advance, which requires a large number of professionals to assist in planning the relationship between the data, and other data can be affected by errors in certain output relationships in the writing process. Because ASP belongs to the stated programming language, the relations are independent, and the correct result output is not affected when irregular errors exist. Since new knowledge and experience are required to be continuously injected into the knowledge base in the later use process, the new knowledge and experience can also conflict with the previous knowledge, but the non-rational rules can not influence the operation of the stable system. The form of demand during use is determined from the beginning of system setup and post-modification is difficult. However, ASP has strong expansibility, and new results can be obtained by adding new constraints to the ASP in the later period, so that the addition in this way cannot conflict with the previous constraints, and on the one hand, in terms of searching constraints, constraint knowledge selection rules can be added to the ASP to search for proper rules.

Meanwhile, the aircraft maintains knowledge theory knowledge, experience knowledge, explicit knowledge and implicit knowledge, some is geometric knowledge, some is non-geometric knowledge, some can be described in a symbol form, and some must be described by an object, so that the aircraft has the characteristics of diversity and dynamics. These all make it difficult for the aircraft maintenance process knowledge to be described in a structured data form. The ASP procedure is compact and easy to handle. If a knowledge-intensive system is written using a conventional procedural or object-oriented programming language (e.g., c++), sufficient knowledge of the data structure is required and more effort is required, with little knowledge base readability. Because of the nature of maintaining knowledge dynamics, it is not possible for a knowledge base programmer to match a real expert implementation in a lower level language, so ASPs are suitable for use with knowledge-intensive application domain modeling issues.

Disclosure of Invention

The invention aims to provide a logic knowledge system for aircraft maintenance, which solves the problem of readability of a traditional process type or object-oriented programming language writing knowledge-intensive system, and can guide researchers to further study the relationship among faults and improve the efficiency of aircraft maintenance tasks.

The technical scheme adopted by the invention is that maintenance knowledge information is collected firstly, and then maintenance evaluation is given, and the method is implemented according to the following steps:

step 1: setting up a knowledge base of the system when the system is built, and adding the knowledge including maintenance planning knowledge and maintenance process knowledge.

Step 2: information about maintenance tasks is collected and maintenance requirements are entered manually by engineers.

Step 3: the clinfo solver performs maintenance behavior solving.

The method comprises the following steps: and managing the maintenance of the aircraft by calculating corresponding maintenance measures according to the solver.

Drawings

FIG. 1 is a general flow chart of the present invention;

FIG. 2 is a knowledge base hierarchical model diagram;

Detailed Description

The invention will be described in detail below with reference to the drawings and the detailed description.

The invention discloses a logic knowledge system for aircraft maintenance, which is implemented by the following steps:

the maintenance knowledge in step 1 can be derived from an aircraft manufacturer or from a first-line maintenance department, so that the aircraft maintenance knowledge can be divided into theoretical knowledge and empirical knowledge. Theoretical knowledge is instructional knowledge that aircraft manufacturer engineers have concluded from regulatory specifications, maintenance manuals, process specifications, and the like. The experience knowledge is the knowledge of maintenance experience, maintenance skills and the like accumulated by the engineers in the first line maintenance department in the maintenance practice of the aircraft. The method is implemented according to the following steps:

step 1.1, knowledge base: in the early stage of designing a system, an aircraft manufacturer establishes a knowledge base according to the expertise of an aircraft, manufacturing, installation and the like, wherein the expertise comprises the expertise of an installation process, a technological process and the like. After delivery, knowledge generated during maintenance is also added to the knowledge base.

Step 1.2, starting from the maintenance process, we propose a maintenance knowledge base hierarchical model as in fig. 2, which is respectively a maintenance main body, a maintenance resource, a maintenance guidance and a maintenance quality confirmation.

Step 2 is performed as follows:

step 2.1, manually inputting maintenance tasks by an engineer: maintenance tasks are classified into scheduled maintenance and unscheduled maintenance. Scheduled maintenance includes timing maintenance based on departure time, time of use, or number of uses, with unscheduled maintenance being based on a particular fault. The maintenance task indicates the detailed content of the maintenance work, including the maintenance time, place, the model and number of the maintenance object, personnel and remarks, and the maintenance task provides the maintenance personnel with basic information of the maintenance task.

Step 3 is performed as follows:

step 3.1, maintenance requirements: aircraft maintenance process knowledge is a data set consisting of maintenance requirements of an aircraft and maintenance process knowledge node associations. Thus, as shown in FIG. 1, the aircraft maintenance process knowledge metadata may be defined as:

maintenance Activities (MA) are derived from the maintenance requirements (MD) and the maintenance knowledge base (MK). Wherein the Maintenance Activities (MA) include maintenance guidelines (MI) and maintenance quality confirmations (MQ). The maintenance requirement consists of a type ID of a maintenance machine type, a specific machine type ID, a maintenance structure ID and the like:

wherein, the Aircraft_Type_Id represents the Type of the Maintenance object, the Aircraft_Id represents the number of the Maintenance object, and the Maintance_Type represents the Maintenance Type, which is generally divided into two main types of planned Maintenance and unplanned Maintenance. The map_part indicates the location of Maintenance. Scheduled maintenance is also known as periodic maintenance, which is typically divided into "machinery days", "6 months", "50 hours", and so on.

Step 3.2, the maintenance knowledge base is a core part of the system, and comprises three parts of Maintenance Object (MO), maintenance Resource (MR), maintenance step (MI) and maintenance quality confirmation (MQ) shown in fig. 2, where mk is a second-order predicate:

1) Maintenance body (MO): the level at which the maintenance agent is located is important and represents the complexity of maintenance. Maintaining predicate structures of a principal:

where object_id is a unique identifier of the maintenance Object, name is a Name of the maintenance Object, up_leave indicates a layer above a layer to which the component belongs, and is used to indicate a membership relationship, leave indicates a layer in which the Object is located, and Ref is an optional remark.

2) Maintenance Resource (MR): the maintenance resource predicates need to give information on the name, number, how to use, etc. of the required maintenance resources. Maintaining the structure of resource predicates:

where Mr_ID represents a unique identifier of a maintenance resource, mo_ID represents a maintenance component that can be performed using the tool, title represents a name of the maintenance resource, describe represents a description of the maintenance resource, ref is an optional remark, typically a note for using the tool or a remark for a hyperlink reference, etc.

3) Maintenance guidelines (MI): the maintenance instruction is the most important part in the maintenance process, and predicates thereof form:

where Mi_ID represents the unique identifier of the Maintenance guide, mo_ID represents the component to which the Maintenance guide applies, title represents the step name, describe represents the representation of the Maintenance step, maintance_steps represents the specific Maintenance step, hyperlinks may be used to represent lengthy Maintenance steps, index is an external link to the key knowledge point of the Maintenance process, and compler represents the creator of the Maintenance guide.

4) Maintenance quality validation (MQ): the maintenance quality confirmation needs to confirm the quality of each maintenance, the predicate construction needs to meet the requirements of different structures, and an operation method for quality confirmation is provided, and the maintenance quality confirmation predicate structure is:

where Mq_ID represents a unique identifier representing a Quality confirmation, mo_ID represents a component corresponding to the maintenance Quality confirmation instance, title represents a name of maintenance Quality confirmation, quality_required is a description of specific requirements for Quality confirmation, test_method represents a method how to check maintenance Quality of the component, and ref is an optional remark.

And 4, carrying out maintenance tasks on the aircraft by engineers according to the maintenance measures calculated by the system.

Claims (5)

1. A logic knowledge system for aircraft maintenance, characterized by being implemented in particular according to the following steps:

step 1: setting a knowledge base of the system when the system is built, and adding maintenance planning knowledge and maintenance process knowledge;

step 2: collecting information of maintenance tasks, and manually inputting maintenance requirements by engineers;

step 3: the clinfo solver carries out maintenance behavior solving;

the method comprises the following steps: and managing the maintenance of the aircraft by calculating corresponding maintenance measures according to the solver.

2. A logic knowledge system for aircraft maintenance according to claim 1, wherein said step 1 is specifically as follows:

step 1.1, knowledge base: in the initial stage of designing a system, an aircraft manufacturer establishes a knowledge base according to the expertise of an aircraft, manufacture and installation, wherein the knowledge base comprises the expertise of an installation flow and a process flow; after delivery, knowledge generated in the maintenance process is also added into the knowledge base;

step 1.2, starting from the maintenance process, we propose a maintenance knowledge base hierarchical model, which is respectively a maintenance main body, a maintenance resource, a maintenance guide and a maintenance quality confirmation.

3. A logic knowledge system for aircraft maintenance according to claim 1, wherein said step 2 is specifically as follows:

step 2.1, manually inputting maintenance tasks by an engineer: maintenance tasks are divided into scheduled maintenance and unscheduled maintenance; scheduled maintenance includes timing maintenance based on departure time, time of use, or number of uses, with unscheduled maintenance based on a specific fault; the maintenance task indicates the detailed content of the maintenance work, including the maintenance time, place, the type and number of the maintenance object, personnel and remarks, and provides basic information of the maintenance task for the maintenance personnel.

4. A logic knowledge system for aircraft maintenance according to claim 1, wherein said step 3 is specifically as follows:

step 3.1, maintenance requirements: the aircraft maintenance process knowledge is a data set formed by the association of maintenance requirements of an aircraft and maintenance process knowledge nodes; thus, as shown in FIG. 1, the aircraft maintenance process knowledge metadata may be defined as:

deriving maintenance behavior (MA) from the maintenance requirements (MD) and the maintenance knowledge base (MK); wherein the Maintenance Activities (MA) include maintenance guidelines (MI) and maintenance quality confirmations (MQ); the maintenance requirement consists of a maintenance model type ID, a specific model ID, a maintenance type ID, and a maintenance structure ID:

wherein, the Aircraft_Type_Id represents the Type of the Maintenance object, the Aircraft_Id represents the number of the Maintenance object, and the Maintance_Type represents the Maintenance Type, which is generally divided into two main types of planned Maintenance and unplanned Maintenance; the Maintenance Part is denoted by the Maintenance Part; scheduled maintenance is also known as periodic maintenance, which is typically divided into "mechanical days", "6 months", "50 hours";

step 3.2, the maintenance knowledge base is a core part of the system, and comprises three parts of Maintenance Object (MO), maintenance Resource (MR), maintenance step (MI) and maintenance quality confirmation (MQ) shown in fig. 2, where mk is a second-order predicate:

1) Maintenance body (MO): the level of maintenance of the main body is important, which represents the complexity of maintenance; maintaining predicate structures of a principal:

wherein object_id is a unique identifier of the maintenance Object, name is a Name of the maintenance Object, up_leave represents a layer above a layer to which the component belongs, is used to represent a membership relationship, leave represents a layer in which the Object is located, and Ref is an optional remark;

2) Maintenance Resource (MR): the maintenance resource predicates need to give information on the names, the numbers and how to use the required maintenance resources; maintaining the structure of resource predicates:

where Mr_ID represents a unique identifier of the maintenance resource, mo_ID represents a maintenance component that can be performed using the tool, title represents the name of the maintenance resource, describe represents a description of the maintenance resource, ref is an optional remark, typically a note for using the tool or a remark for hyperlink reference;

3) Maintenance guidelines (MI): the maintenance instruction is the most important part in the maintenance process, and predicates thereof form:

where Mi_ID represents the unique identifier of the Maintenance guide, mo_ID represents the component to which the Maintenance guide applies, title represents the step name, describe represents the representation of the Maintenance step, maintance_steps represents the specific Maintenance step, hyperlinks may be used to represent lengthy Maintenance steps, index is an external link to the key knowledge point of the Maintenance process, and compler represents the creator of the Maintenance guide;

4) Maintenance quality validation (MQ): the maintenance quality confirmation needs to confirm the quality of each maintenance, the predicate construction needs to meet the requirements of different structures, and an operation method for quality confirmation is provided, and the maintenance quality confirmation predicate structure is:

where Mq_ID represents a unique identifier representing a Quality confirmation, mo_ID represents a component corresponding to the maintenance Quality confirmation instance, title represents a name of maintenance Quality confirmation, quality_required is a description of specific requirements for Quality confirmation, test_method represents a method how to check maintenance Quality of the component, and ref is an optional remark.

5. A logic knowledge system for aircraft maintenance according to claim 1, wherein said step 4 is specifically as follows: and the engineer performs maintenance tasks on the aircraft according to the maintenance measures calculated by the system.

Images