Proposed Case-based Reasoning System Research Articles

Case-based reasoning system for fault diagnosis of aero-engines

Fault diagnosis in aero-engines typically requires an experienced expert for understanding and detecting the cause of faults. However, accurate and quick identification of fault parts is difficult for maintenance crews owing to the complexity of aero-engines. In this study, we developed a case-based reasoning (CBR) system with a highly accurate novel similarity measure for fault diagnosis of aero-engines by retrieving similar fault cases. The proposed CBR system is established based on 143 cases with the knowledge of correctly diagnosed and successfully resolved aero-engine faults, which constitutes the first tentative case base in the field of aero-engine fault diagnosis. The proposed case similarity measure for fault diagnosis of aero-engines integrates three local similarity measures associated with different attributes, especially among which a tree-based semantic similarity measure is proposed to define the relationship between the fault part and fault mode based on a semantic graph incorporated into the aero-engine tree structure. The proposed CBR system is evaluated using the k-nearest neighbors algorithm and 5-fold cross-validation. The system exhibited high retrieval accuracy with all cases collected from real-world scenarios of aero-engine fault diagnosis. Our study shows great promise that the experience-based decisions yielded from the results can aid in aero-engine maintenance and support services.

Synchronized measurements-based wide-area static security assessment and classification of power systems using case based reasoning classifiers

This paper proposes a human intuition-based concept of Case-Based Reasoning (CBR) for Static Security Assessment (SSA) of power systems. Prior to security assessment, a composite Security Index (SI) defined in terms of line overload and voltage violation, is computed for each training pattern to label them into either secure or insecure class. Out of all phasor measurements obtained using Phasor Measurement Units (PMUs), only the real and reactive power flows, which are identified as relevant features are given as inputs to the proposed CBR system. The proposed approach has been implemented on IEEE 14 bus, IEEE 30 bus and Indian 246 bus networks and the results have been compared with those obtained using Support Vector Machine (SVM) and Artificial Neural Network (ANN) based classifiers. The results indicate that the proposed approach is more efficient, reliable and outperforms the above said classifiers in terms of classification accuracy at a comparable computation cost.

Application of CBR systems in the process of energy retrofit of single-family detached houses

In Poland, there are more than 5 million single-family detached houses, and 35% among them are non-insulated. In the article, we present a concept of application of the case-based reasoning (CBR) systems to support the process of energy retrofit of single-family detached houses. In the proposed CBR system, the energy retrofit process was described using parameters with text and numeric values and also using linguistic variables defined using fuzzy sets. If an investor, designer or contractor encounters a problem in the process of the energy retrofit, he may look for a solution to this problem or similar problems in the CBR system. If he does not find a solution and creates it himself or modifies an existing one, he should put it in the CBR system so that others can benefit from his experience. The proposed CBR system will help in solving the basic Polish (but also Eastern European) problem of energy-saving constructions which is the lack of adequate knowledge of participants in the investment process.

Handling missing values and unmatched features in a CBR system for hydro-generator design

Hydro-generator design is a complex problem and case based reasoning (CBR) can improve its efficiency, but there are missing values and unmatched features which decrease the accuracy of CBR. In order to solve the problems brought by missing values and unmatched features, a similarity measurement is proposed by improving the edit distance which is widely used as a similarity measurement. In the proposed CBR system, the case base is constructed based on domain ontology to improve the retrieval efficiency. Then a case representation is proposed and cases are represented by a unified tree model. Next, by combining the edit distance with feature weights and the semantic meanings of case nodes, the cost function is proposed to measure the semantic difference and the conditions which make it a metric are discussed. Lastly, the similarity function is defined based on the cost function. A case study is presented to illustrate the use of the proposed CBR system, and then the experiments are executed to evaluate its performance in dealing with missing values and unmatched features respectively. The results validate that the proposed CBR system can handle missing values and unmatched features effectively.

A CBR system for injection mould design based on ontology: A case study

Case-based reasoning (CBR) often shows significant promise for improving the effectiveness of design support in mould design, which is a domain strong in practice but poor in theory. However, existing CBR systems lack semantic understanding, which is important for intelligent knowledge retrieval in design support system. This hinders the application of CBR in injection mould design. In order to develop an intelligent CBR system and meet the need of design support for injection mould design, this paper integrates ontology technology into a CBR system by constructing domain ontology as case-base with a new method, in which two means of acquisition are combined, one is acquiring ontology from existing ontologies, the other from established engineering knowledge resources, and proposing a new semantic retrieval method as the first grade case retrieval. Numerical measurement is also employed as the second grade case retrieval, which adopts various methods to calculate different types of attribute values. A case is executed to illustrate the use of proposed CBR system, then a lot of experiments are organized to evaluate its performance and the result shows that the proposed approach outperforms existing CBR systems.

A novel CBR system for numeric prediction

Case-based reasoning (CBR) solves new problems by recalling and reusing the solutions to similar problems. Despite its popularity and simplicity, relatively little work has been done to improve CBR for numeric prediction. To predict numeric values accurately and efficiently, this paper develops a novel case indexing approach and a simple attribute weighting method for CBR. This study evaluates the proposed CBR system using nine well-known data sets, showing that it achieves better efficiency and accuracy than conventional CBR. This study also applies the proposed CBR system to solve the due date assignment (DDA) problem in a dynamic wafer fabrication factory to determine if it’s expected benefits can be observed in practice. Experimental results show that the proposed CBR system significantly improves job due date prediction.

A novel case based reasoning approach to radiotherapy planning

Radiotherapy planning is a complex problem which requires both expertise and experience of an oncologist. A case based reasoning (CBR) system is developed to generate dose plans for prostate cancer patients. The proposed approach captures the expertise and experience of oncologists in treating previous patients and recommends a dose in phase I and phase II of the treatment of a new patient considering also the success rate of the treatment. The proposed CBR system employs a modified Dempster–Shafer theory to fuse dose plans suggested by the most similar cases retrieved from the case base. In order to mimic the continuous learning characteristic of oncologists, the weights corresponding to each feature used in the retrieval process are updated automatically each time after generating a treatment plan for a new patient. The efficiency of the proposed methodology has been validated using real data sets collected from the Nottingham University Hospitals NHS, City Hospital Campus, UK. Experiments demonstrated that for most of the patients, the dose plan generated by our approach is coherent with the dose plan suggested by an experienced oncologist. This methodology can assist both new and experienced oncologists in the treatment planning.

A forecasting solution to the oil spill problem based on a hybrid intelligent system

Oil spills represent one of the most destructive environmental disasters. Predicting the possibility of finding oil slicks in a certain area after an oil spill can be critical in reducing environmental risks. The system presented here uses the Case-Based Reasoning (CBR) methodology to forecast the presence or absence of oil slicks in certain open sea areas after an oil spill. CBR is a computational methodology designed to generate solutions to certain problems by analysing previous solutions given to previously solved problems. The proposed CBR system includes a novel network for data classification and retrieval. This type of network, which is constructed by using an algorithm to summarize the results of an ensemble of Self-Organizing Maps, is explained and analysed in the present study. The Weighted Voting Superposition (WeVoS) algorithm mainly aims to achieve the best topographically ordered representation of a dataset in the map. This study shows how the proposed system, called WeVoS-CBR, uses information such as salinity, temperature, pressure, number and area of the slicks, obtained from various satellites to accurately predict the presence of oil slicks in the north-west of the Galician coast, using historical data.

A CBR method for CFW prevention and treatment

Different from the conventional methods, this paper presents a case-based reasoning (CBR) method for cucumber fusarium wilt (CFW) prevention and treatment, with vantage case-based indexing and retrieval mechanism, sensitivity analysis based reasoning method and Recall evaluation based case retaining strategy. The optimal interval of case dissimilarity distance threshold T is inferred from the experimental results by X fold cross-validation method. The experiments and application show that useful information can be generated for CFW prevention and treatment and the dynamic adjustment for decision-making of cucumber planting can be assisted by the proposed CBR system.

Reducing the Memory Size of a Fuzzy Case-Based Reasoning System Applying Rough Set Techniques

Early work on case-based reasoning (CBR) reported in the literature shows the importance of soft computing techniques applied to different stages of the classical four-step CBR life cycle. This correspondence proposes a reduction technique based on rough sets theory capable of minimizing the case memory by analyzing the contribution of each case feature. Inspired by the application of the minimum description length principle, the method uses the granularity of the original data to compute the relevance of each attribute. The rough feature weighting and selection method is applied as a preprocessing step prior to the generation of a fuzzy rule system, which is employed in the revision phase of the proposed CBR system. Experiments using real oceanographic data show that the rough sets reduction method maintains the accuracy of the employed fuzzy rules, while reducing the computational effort needed in its generation and increasing the explanatory strength of the fuzzy rules

Demonstration of a tool for automatic learning and re-use of knowledge in the activated sludge process

Wastewater treatment plant operators encounter complex operational problems related to the activated sludge process and usually respond to these by applying their own intuition and by taking advantage of what they have learnt from past experiences of similar problems. However, previous process experiences are not easy to integrate in numerical control, and new tools must be developed to enable re-use of plant operating experience. The aim of this paper is to investigate the usefulness of a case-based reasoning (CBR) approach to apply learning and re-use of knowledge gained during past incidents to confront actual complex problems through the IWA/COST Benchmark protocol. A case study shows that the proposed CBR system achieves a significant improvement of the benchmark plant performance when facing a high-flow event disturbance.

Intelligent aircraft maintenance support system using genetic algorithms and case-based reasoning

The maintenance of aircraft components is crucial for avoiding aircraft accidents and aviation fatalities. To provide reliable and effective maintenance support, it is important for the airline companies to utilise previous repair experience with the aid of advanced decision support technology. Case-based reasoning (CBR) is a machine learning method that adapts previous similar cases to solve current problems. To effectively retrieve similar aircraft maintenance cases, this research proposes using a CBR system to aid electronic ballast fault diagnosis of Boeing 747-400 airplanes. By employing genetic algorithms (GA) to enhance dynamic weighting and the design of non-similarity functions, the proposed CBR system is able to achieve superior learning performance as compared to those with either equal/varied weights or linear similarity functions.